1<!DOCTYPE html> 2<html lang="en-US"> 3 <head> 4 <meta charset="utf-8"> 5 <meta http-equiv="x-ua-compatible" content="ie=edge"> 6 <meta name="viewport" content="width=device-width, initial-scale=1"> 7 <meta name="application-name" content="Zhan Wang"> 8 <meta name="theme-color" content="#fba211"> 9 <link rel="shortcut icon" href="/favicon.png"/> 10 <link rel="icon" type="image/png" href="/favicon.png" sizes="250x250" /> 11 12 <!-- Begin Jekyll SEO tag v2.8.0 --> 13<title>Publications | Zhan Wang</title> 14<meta name="generator" content="Jekyll v3.10.0" /> 15<meta property="og:title" content="Publications" /> 16<meta property="og:locale" content="en_US" /> 17<meta name="description" content="The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University." /> 18<meta property="og:description" content="The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University." /> 19<link rel="canonical" href="https://zhanwang071.github.io//publications/" /> 20<meta property="og:url" content="https://zhanwang071.github.io//publications/" /> 21<meta property="og:site_name" content="Zhan Wang" /> 22<meta property="og:type" content="website" /> 23<meta name="twitter:card" content="summary" /> 24<meta property="twitter:title" content="Publications" /> 25<meta name="twitter:site" content="@zoeywuwa" />
26<script type="application/ld+json"> 27{"@context":"https://schema.org","@type":"WebPage","description":"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.","headline":"Publications","url":"https://zhanwang071.github.io//publications/"}</script>
27 28<!-- End Jekyll SEO tag --> 29 30 31 <link rel="alternate" type="application/rss+xml" title="Zhan Wang" href="https://zhanwang071.github.io//feed.xml"> 32 33 <link href="https://use.fontawesome.com/releases/v6.2.1/css/all.css" rel="stylesheet"> 34 <link href="/styles.css" rel="stylesheet"> 35</head> 36 37 <body> 38 39<header class="page-header"> 40 <nav class="container"> 41 <a class="site-title" href="/">Zhan Wang</a> 42 43 <a href="/projects/">Projects</a> 44 <a href="/publications/" aria-current="page">Publications</a> 45 <!-- <a href="/talks/" >Talks</a> --> 46 <!-- <a href="/blog/" >Blog</a> --> 47 <a href="/cv/">CV</a> 48 49 <span class="external"> 50 <!-- <a rel="me" href="https://vis.social/@dom"><i class="fab fa-mastodon" aria-hidden="true"></i> Mastodon</a> --> 51 <a href="https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en"><i class="fab fa-google" aria-hidden="true"></i> Scholar</a> 52 <a href="https://github.com/zhanwang071"><i class="fab fa-github" aria-hidden="true"></i> GitHub</a> 53 </span> 54 </nav> 55</header> 56 57 58 <div class="page-content"> 59 <section class="container pubs"> 60 <h1 class="hidden" id="publications">Publications</h1> 61 62<!-- <div id="facets" class="hidden"> 63 <div class="facet" id="venue_tags"> 64 <strong>Venue</strong> 65 <ul></ul> 66 </div> 67 <div class="facet" id="authors"> 68 <strong>Author</strong> 69 <ul></ul> 70 </div> 71 <div class="facet" id="tags"> 72 <strong>Tag</strong> 73 <ul></ul> 74 </div> 75 <div class="facet" id="type"> 76 <strong>Type</strong> 77 <ul></ul> 78 </div> 79 <div class="facet" id="awards"> 80 <strong>Award</strong> 81 <ul></ul> 82 </div> 83 84</div> 85 86<div class="p1 db"> 87 <input id="ft-search" type="search" placeholder="Search papers..." /> 88</div> 89 90<label id="only-highlight" class="hidden"> 91 <input type="checkbox" id="highlight"> 92 Show only highlights 93</label> --> 94 95<!-- <p id="clear-filters" class="hidden"> 96 <i class="fas fa-times-circle" aria-hidden="true"></i> Clear all filters. <span id="count_hidden">X</span> of <span id="count_total">X</span> publications are hidden by the filters. 97</p> --> 98 99<h2 id="y2026" class="year">2026</h2> 100 101<div class="publication" data-pub="{"path":"_publications/2026-03-less.html","relative_path":"_publications/2026-03-less.html","excerpt":"","previous":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","excerpt":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","previous":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","excerpt":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","previous":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","id":"/publications/2025-01-cinefolio","collection":"publications","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["
101Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"id":"/publications/2025-02-vizta","collection":"publications","next":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","id":"/publications/2025-03-drama","collection":"publications","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<meta property=\"og:description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\",\"headline\":\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\",\"url\":\"https://zhanwang071.github.io//publications/2025-02-vizta\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
101 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"id":"/publications/2025-03-drama","collection":"publications","next":{"path":"_publications/2026-03-less.html","relative_path":"_publications/2026-03-less.html","excerpt":"","previous":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","id":"/publications/2025-03-drama","collection":"publications","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"id":"/publications/2026-03-less","collection":"publications","next":null,"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\" />\n<meta name=\"author\" content=\"Zhiyao Yang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2026-03-less\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2026-03-less\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhiyao Yang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhiyao Yang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\",\"url\":\"https://zhanwang071.github.io//publications/2026-03-less\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
101 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2026-03-less","draft":false,"categories":[],"layout":"publication","year":2026,"title":"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.","authors":["Zhiyao Yang","Zhan Wang","Yihou Han","Qian Zhu","Wei Zeng"],"type":["Journal"],"venue":"Virtual Reality","venue_tags":["VR"],"tags":["Responsive Design","Data Visualization","Virtual Reality"],"highlight":false,"slug":"2026-03-less","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"author\" content=\"Jian Yu\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<meta property=\"og:description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
101igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Jian Yu\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Jian Yu\"},\"description\":\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\",\"headline\":\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\",\"url\":\"https://zhanwang071.github.io//publications/2025-03-drama\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
101igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"id":"/publications/2026-03-less","collection":"publications","next":null,"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\" />\n<meta name=\"author\" content=\"Zhiyao Yang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2026-03-less\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2026-03-less\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhiyao Yang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhiyao Yang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\",\"url\":\"https://zhanwang071.github.io//publications/2026-03-less\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
101 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2026-03-less","draft":false,"categories":[],"layout":"publication","year":2026,"title":"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.","authors":["Zhiyao Yang","Zhan Wang","Yihou Han","Qian Zhu","Wei Zeng"],"type":["Journal"],"venue":"Virtual Reality","venue_tags":["VR"],"tags":["Responsive Design","Data Visualization","Virtual Reality"],"highlight":false,"slug":"2026-03-less","ext":".html"}"> 102 <h3 id="/publications/2026-03-less">Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.</h3> 103 <div class="authors"> 104 Zhiyao Yang, <b>Zhan Wang</b>, Yihou Han, Qian Zhu, Wei Zeng. 105 </div> 106 107 <div class="venue"> 108 109 Virtual Reality<!-- 110 111 --> 112 2026 113 </div> 114 115 116 117 118 119 120 121 122 123 124</div> 125 126<h2 id="y2025" class="year">2025</h2> 127 128<div class="publication" data-pub="{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","excerpt":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","previous":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","excerpt":"","previous":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","excerpt":"","previous":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","id":"/publications/2024-02-gen4ai","collection":"publications","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["
128Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"id":"/publications/2024-03-spatiallearning","collection":"publications","next":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","id":"/publications/2024-04-isea","collection":"publications","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"author\" content=\"Yifei Zhang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yifei Zhang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yifei Zhang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\",\"url\":\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
128 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"id":"/publications/2024-04-isea","collection":"publications","next":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","excerpt":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","previous":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","id":"/publications/2024-04-isea","collection":"publications","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"id":"/publications/2025-01-cinefolio","collection":"publications","next":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","id":"/publications/2025-02-vizta","collection":"publications","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["
128Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<meta property=\"og:description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
128sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\",\"headline\":\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\",\"url\":\"https://zhanwang071.github.io//publications/2025-01-cinefolio\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
128sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["
128Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Staying vigilant in the Age of AI: From content generation to content authentication. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"author\" content=\"Yufan Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yufan Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yufan Li\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Staying vigilant in the Age of AI: From content generation to content authentication.\",\"url\":\"https://zhanwang071.github.io//publications/2024-04-isea\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
128 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"id":"/publications/2025-01-cinefolio","collection":"publications","next":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","excerpt":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","previous":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","excerpt":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","previous":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","id":"/publications/2024-04-isea","collection":"publications","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"id":"/publications/2025-01-cinefolio","collection":"publications","next":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","id":"/publications/2025-02-vizta","collection":"publications","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["
128Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<meta property=\"og:description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
128sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\",\"headline\":\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\",\"url\":\"https://zhanwang071.github.io//publications/2025-01-cinefolio\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
128sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["
128Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"id":"/publications/2025-02-vizta","collection":"publications","next":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","excerpt":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","previous":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","id":"/publications/2025-02-vizta","collection":"publications","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"id":"/publications/2025-03-drama","collection":"publications","next":{"path":"_publications/2026-03-less.html","relative_path":"_publications/2026-03-less.html","id":"/publications/2026-03-less","collection":"publications","url":"/publications/2026-03-less","draft":false,"categories":[],"layout":"publication","year":2026,"title":"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.","authors":["Zhiyao Yang","Zhan Wang","Yihou Han","Qian Zhu","Wei Zeng"],"type":["
128Journal"],"venue":"Virtual Reality","venue_tags":["VR"],"tags":["Responsive Design","Data Visualization","Virtual Reality"],"highlight":false,"slug":"2026-03-less","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"author\" content=\"Jian Yu\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<meta property=\"og:description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
128igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Jian Yu\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Jian Yu\"},\"description\":\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\",\"headline\":\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\",\"url\":\"https://zhanwang071.github.io//publications/2025-03-drama\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
128igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<meta property=\"og:description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\",\"headline\":\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\",\"url\":\"https://zhanwang071.github.io//publications/2025-02-vizta\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
128 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
128sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<meta property=\"og:description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
128sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\",\"headline\":\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\",\"url\":\"https://zhanwang071.github.io//publications/2025-01-cinefolio\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
128 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"}"> 129 <h3 id="/publications/2025-01-cinefolio">CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.</h3> 130 <div class="authors"> 131 <b>Zhan Wang</b>, Qian Zhu, David Yip, Fugee Tsung, Wei Zeng. 132 </div> 133 134 <div class="venue"> 135 136 Visual Informatics<!-- 137 138 --> 139 2025 140 </div> 141 142 143 144 145 146 147 148 149 150 151 <div class="extra-links"> 152 153 <a href="http://doi.org/10.1016/j.visinf.2025.100259"> 154 <i class="fas fa-anchor" aria-hidden="true"></i> DOI 155 </a> 156 157 158 <a href="/papers/2025-cinefolio.pdf"> 159 <i class="far fa-file-pdf" aria-hidden="true"></i> PDF 160 </a> 161 162 163 <!-- --> 164 165 166 167 168 169 170 <!-- --> 171 172 173 </div> 174 175</div> 176 177<div class="publication" data-pub="{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","excerpt":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","previous":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","excerpt":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
177sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","previous":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","excerpt":"","previous":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","id":"/publications/2024-03-spatiallearning","collection":"publications","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"id":"/publications/2024-04-isea","collection":"publications","next":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","id":"/publications/2025-01-cinefolio","collection":"publications","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Staying vigilant in the Age of AI: From content generation to content authentication. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"author\" content=\"Yufan Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yufan Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yufan Li\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Staying vigilant in the Age of AI: From content generation to content authentication.\",\"url\":\"https://zhanwang071.github.io//publications/2024-04-isea\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
177 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"id":"/publications/2025-01-cinefolio","collection":"publications","next":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","excerpt":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","previous":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","id":"/publications/2025-01-cinefolio","collection":"publications","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"id":"/publications/2025-02-vizta","collection":"publications","next":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","id":"/publications/2025-03-drama","collection":"publications","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["
177Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<meta property=\"og:description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\",\"headline\":\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\",\"url\":\"https://zhanwang071.github.io//publications/2025-02-vizta\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
177 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
177sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<meta property=\"og:description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
177sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\",\"headline\":\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\",\"url\":\"https://zhanwang071.github.io//publications/2025-01-cinefolio\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
177 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"id":"/publications/2025-02-vizta","collection":"publications","next":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","excerpt":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
177igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","previous":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","excerpt":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","previous":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","id":"/publications/2025-01-cinefolio","collection":"publications","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"id":"/publications/2025-02-vizta","collection":"publications","next":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","id":"/publications/2025-03-drama","collection":"publications","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<meta property=\"og:description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\",\"headline\":\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\",\"url\":\"https://zhanwang071.github.io//publications/2025-02-vizta\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
177 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"id":"/publications/2025-03-drama","collection":"publications","next":{"path":"_publications/2026-03-less.html","relative_path":"_publications/2026-03-less.html","excerpt":"","previous":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","id":"/publications/2025-03-drama","collection":"publications","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"id":"/publications/2026-03-less","collection":"publications","next":null,"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\" />\n<meta name=\"author\" content=\"Zhiyao Yang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2026-03-less\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2026-03-less\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhiyao Yang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhiyao Yang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\",\"url\":\"https://zhanwang071.github.io//publications/2026-03-less\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
177 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2026-03-less","draft":false,"categories":[],"layout":"publication","year":2026,"title":"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.","authors":["Zhiyao Yang","Zhan Wang","Yihou Han","Qian Zhu","Wei Zeng"],"type":["Journal"],"venue":"Virtual Reality","venue_tags":["VR"],"tags":["Responsive Design","Data Visualization","Virtual Reality"],"highlight":false,"slug":"2026-03-less","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"author\" content=\"Jian Yu\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<meta property=\"og:description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
177igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Jian Yu\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Jian Yu\"},\"description\":\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\",\"headline\":\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\",\"url\":\"https://zhanwang071.github.io//publications/2025-03-drama\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
177igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<meta property=\"og:description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\",\"headline\":\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\",\"url\":\"https://zhanwang071.github.io//publications/2025-02-vizta\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
177 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"}"> 178 <h3 id="/publications/2025-02-vizta">VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.</h3> 179 <div class="authors"> 180 Liangwei Wang, <b>Zhan Wang</b>, Shishi Xiao, Le Liu, Fugee Tsung, Wei Zeng. 181 </div> 182 183 <div class="venue"> 184 185 Computer Graphics Forum<!-- 186 187 --> 188 2025 189 </div> 190 191 192 193 194 195 196 197 198 199 200</div> 201 202<div class="publication" data-pub="{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","excerpt":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
202igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","previous":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","excerpt":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","previous":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","excerpt":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","previous":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","id":"/publications/2024-04-isea","collection":"publications","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"id":"/publications/2025-01-cinefolio","collection":"publications","next":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","id":"/publications/2025-02-vizta","collection":"publications","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["
202Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<meta property=\"og:description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
202sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\",\"headline\":\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\",\"url\":\"https://zhanwang071.github.io//publications/2025-01-cinefolio\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
202sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["
202Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"id":"/publications/2025-02-vizta","collection":"publications","next":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","excerpt":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","previous":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","id":"/publications/2025-02-vizta","collection":"publications","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"id":"/publications/2025-03-drama","collection":"publications","next":{"path":"_publications/2026-03-less.html","relative_path":"_publications/2026-03-less.html","id":"/publications/2026-03-less","collection":"publications","url":"/publications/2026-03-less","draft":false,"categories":[],"layout":"publication","year":2026,"title":"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.","authors":["Zhiyao Yang","Zhan Wang","Yihou Han","Qian Zhu","Wei Zeng"],"type":["
202Journal"],"venue":"Virtual Reality","venue_tags":["VR"],"tags":["Responsive Design","Data Visualization","Virtual Reality"],"highlight":false,"slug":"2026-03-less","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"author\" content=\"Jian Yu\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<meta property=\"og:description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
202igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Jian Yu\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Jian Yu\"},\"description\":\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\",\"headline\":\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\",\"url\":\"https://zhanwang071.github.io//publications/2025-03-drama\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
202igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<meta property=\"og:description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\",\"headline\":\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\",\"url\":\"https://zhanwang071.github.io//publications/2025-02-vizta\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
202 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"id":"/publications/2025-03-drama","collection":"publications","next":{"path":"_publications/2026-03-less.html","relative_path":"_publications/2026-03-less.html","excerpt":"","previous":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","excerpt":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","previous":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","id":"/publications/2025-02-vizta","collection":"publications","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["
202Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"id":"/publications/2025-03-drama","collection":"publications","next":{"path":"_publications/2026-03-less.html","relative_path":"_publications/2026-03-less.html","id":"/publications/2026-03-less","collection":"publications","url":"/publications/2026-03-less","draft":false,"categories":[],"layout":"publication","year":2026,"title":"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.","authors":["Zhiyao Yang","Zhan Wang","Yihou Han","Qian Zhu","Wei Zeng"],"type":["Journal"],"venue":"Virtual Reality","venue_tags":["VR"],"tags":["Responsive Design","Data Visualization","Virtual Reality"],"highlight":false,"slug":"2026-03-less","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"author\" content=\"Jian Yu\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<meta property=\"og:description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
202igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Jian Yu\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Jian Yu\"},\"description\":\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\",\"headline\":\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\",\"url\":\"https://zhanwang071.github.io//publications/2025-03-drama\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
202igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"id":"/publications/2026-03-less","collection":"publications","next":null,"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\" />\n<meta name=\"author\" content=\"Zhiyao Yang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2026-03-less\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2026-03-less\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhiyao Yang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhiyao Yang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.\",\"url\":\"https://zhanwang071.github.io//publications/2026-03-less\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
202 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2026-03-less","draft":false,"categories":[],"layout":"publication","year":2026,"title":"Less is More Afar: Responsive Design for Situated Visualization Based on Viewing Distance.","authors":["Zhiyao Yang","Zhan Wang","Yihou Han","Qian Zhu","Wei Zeng"],"type":["Journal"],"venue":"Virtual Reality","venue_tags":["VR"],"tags":["Responsive Design","Data Visualization","Virtual Reality"],"highlight":false,"slug":"2026-03-less","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"author\" content=\"Jian Yu\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<meta property=\"og:description\" content=\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
202igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-03-drama\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Jian Yu\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Jian Yu\"},\"description\":\"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiencesâ aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\",\"headline\":\"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.\",\"url\":\"https://zhanwang071.github.io//publications/2025-03-drama\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their al
202igned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"The excavation of intangible cultural heritage (ICH) contextual information carried by cultural heritage (CH) presents new possibilities with the continuous advancement of digital technology. However, when visiting CH sites, due to physical limitations or conservation principles, audiences often focus only on the tangible aspects of CH, while neglecting or struggling to interpret the intangible cultural background. Therefore, presenting the rich intangible cultural contextual information associated with tangible CH is growing in significance, and virtual display methods are gaining prominence. In this work, we select the Shiwan pottery sculptures from over 100 years ago, located in the Chen Clan Ancestral Hall in China, as the subject of our CH research. We conduct a narrative analysis of the intended interpretation of its historical storytelling context. Based on this, we design a new system for embedding ICH narrative information into tangible CH, enabling their aligned comprehension, using 3D scene construction technology, motion capture and virtual reality (VR) technology. The system reimagines the storyline and spatial theatrical scene of Chinese local drama (Cantonese Opera) as ICH in the tangible CH. Results of a comparative user experiment show that the VR immersive interactive system enhances audiences' aligned understanding of ICH contextual storytelling when visiting CH, and it generates a deep interest in ICH.","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"}"> 203 <h3 id="/publications/2025-03-drama">Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.</h3> 204 <div class="authors"> 205 Jian Yu, <b>Zhan Wang</b>, Yifan Cao, Hao Cui, Wei Zeng. 206 </div> 207 208 <div class="venue"> 209
210 Journal on Computing and Cultural Heritage<!-- 211 212 --> 213 2025 214 </div> 215 216 217 218 219 220 221 222 223 224 225</div> 226 227<h2 id="y2024" class="year">2024</h2> 228 229<div class="publication" data-pub="{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","excerpt":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","previous":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","excerpt":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","previous":{"path":"_publications/2023-01-geocamera.html","relative_path":"_publications/2023-01-geocamera.html","excerpt":"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.","previous":null,"id":"/publications/2023-01-geocamera","collection":"publications","next":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","id":"/publications/2023-02-storymap","collection":"publications","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["
229Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\" />\n<meta name=\"author\" content=\"Wenchao Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\" />\n<meta property=\"og:description\" content=\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-01-geocamera\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-01-geocamera\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Wenchao Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Wenchao Li\"},\"description\":\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\",\"headline\":\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\",\"url\":\"https://zhanwang071.github.io//publications/2023-01-geocamera\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
229 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.","url":"/publications/2023-01-geocamera","draft":false,"categories":[],"layout":"publication","year":2023,"title":"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.","authors":["Wenchao Li","Zhan Wang","Yun Wang","Di Weng","Liwenhan Xie","Siming Chen","Haidong Zhang","Huamin Qu"],"type":["Conference"],"venue":"Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems","venue_location":"Hamburg, Germany","venue_tags":["CHI"],"tags":["Human-centered computing","Interactive systems and tools"],"venue_url":"https://chi2023.acm.org/","arxiv":"2303.06460","short_doi":"10.1145/3544548.3581470","highlight":true,"pdf":"https://dl.acm.org/doi/pdf/10.1145/3544548.3581470","video":"https://www.youtube.com/watch?v=SFBmBSHjKpA","slug":"2023-01-geocamera","ext":".html"},"id":"/publications/2023-02-storymap","collection":"publications","next":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","excerpt":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","previous":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","id":"/publications/2023-02-storymap","collection":"publications","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["
229Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"id":"/publications/2024-01-virtuwander","collection":"publications","next":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","id":"/publications/2024-02-gen4ai","collection":"publications","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<meta property=\"og:description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\",\"headline\":\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\",\"url\":\"https://zhanwang071.github.io//publications/2024-01-virtuwander\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
229 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<meta property=\"og:description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\",\"headline\":\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\",\"url\":\"https://zhanwang071.github.io//publications/2023-02-storymap\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
229 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"id":"/publications/2024-01-virtuwander","collection":"publications","next":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","excerpt":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
229nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","previous":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","excerpt":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","previous":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","id":"/publications/2023-02-storymap","collection":"publications","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"id":"/publications/2024-01-virtuwander","collection":"publications","next":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","id":"/publications/2024-02-gen4ai","collection":"publications","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<meta property=\"og:description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\",\"headline\":\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\",\"url\":\"https://zhanwang071.github.io//publications/2024-01-virtuwander\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
229 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"id":"/publications/2024-02-gen4ai","collection":"publications","next":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","excerpt":"","previous":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","id":"/publications/2024-02-gen4ai","collection":"publications","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"id":"/publications/2024-03-spatiallearning","collection":"publications","next":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","id":"/publications/2024-04-isea","collection":"publications","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"author\" content=\"Yifei Zhang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yifei Zhang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yifei Zhang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\",\"url\":\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
229 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Generative AI for visualization: State of the art and future directions. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"author\" content=\"Yilin Ye\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By
229summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<meta property=\"og:description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yilin Ye\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yilin Ye\"},\"description\":\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
229nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\",\"headline\":\"Generative AI for visualization: State of the art and future directions.\",\"url\":\"https://zhanwang071.github.io//publications/2024-02-gen4ai\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
229 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<meta property=\"og:description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\",\"headline\":\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\",\"url\":\"https://zhanwang071.github.io//publications/2024-01-virtuwander\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
229 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"}"> 230 <h3 id="/publications/2024-01-virtuwander">VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.</h3> 231 <div class="authors"> 232 <b>Zhan Wang</b>, Linping Yuan, Liangwei Wang, Bingchuan Jiang, Wei Zeng. 233 </div> 234 235 <div class="venue"> 236 237 <a href="https://chi2024.acm.org/"> 238 239 Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems<!-- 240 241 --> in Hawaii, US<!-- 242 243 --></a> 244 2024 245 </div> 246 247 248 249 250 251 252 253 254 255 256 <div class="extra-links"> 257 258 <a href="http://doi.org/10.1145/3613904.3642235"> 259 <i class="fas fa-anchor" aria-hidden="true"></i> DOI 260 </a> 261 262 263 <a href="https://arxiv.org/pdf/2401.11923.pdf"> 264 <i class="far fa-file-pdf" aria-hidden="true"></i> PDF 265 </a> 266 267 268 <!-- 269 <a href="https://ar5iv.org/html/2401.11923"> 270 <i class="fab fa-html5" aria-hidden="true"></i> arXiv 271 </a> 272 --> 273 274 275 <a href="https://arxiv.org/abs/2401.11923"> 276 <i class="fas fa-archive" aria-hidden="true"></i> arXiv 277 </a> 278 279 280 <a href="https://youtu.be/Dt3g4NMFmUc"> 281 <i class="fas fa-film" aria-hidden="true"></i> Video 282 </a> 283 284 285 286 287 <!-- 288 <a href="https://arxiv2bibtex.org/?q=2401.11923"> 289 <i class="fas fa-book" aria-hidden="true"></i> Bibtex 290 </a> 291 --> 292 293 294 </div> 295 296</div> 297 298<div class="publication" data-pub="{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","excerpt":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
298nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","previous":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","excerpt":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","previous":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","excerpt":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","previous":{"path":"_publications/2023-01-geocamera.html","relative_path":"_publications/2023-01-geocamera.html","id":"/publications/2023-01-geocamera","collection":"publications","url":"/publications/2023-01-geocamera","draft":false,"categories":[],"layout":"publication","year":2023,"title":"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.","authors":["Wenchao Li","Zhan Wang","Yun Wang","Di Weng","Liwenhan Xie","Siming Chen","Haidong Zhang","Huamin Qu"],"type":["Conference"],"venue":"Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems","venue_location":"Hamburg, Germany","venue_tags":["CHI"],"tags":["Human-centered computing","Interactive systems and tools"],"venue_url":"https://chi2023.acm.org/","arxiv":"2303.06460","short_doi":"10.1145/3544548.3581470","highlight":true,"pdf":"https://dl.acm.org/doi/pdf/10.1145/3544548.3581470","video":"https://www.youtube.com/watch?v=SFBmBSHjKpA","slug":"2023-01-geocamera","ext":".html"},"id":"/publications/2023-02-storymap","collection":"publications","next":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","id":"/publications/2024-01-virtuwander","collection":"publications","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<meta property=\"og:description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\",\"headline\":\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\",\"url\":\"https://zhanwang071.github.io//publications/2023-02-storymap\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n\n</section
298>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"id":"/publications/2024-01-virtuwander","collection":"publications","next":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","excerpt":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
298nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","previous":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","id":"/publications/2024-01-virtuwander","collection":"publications","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"id":"/publications/2024-02-gen4ai","collection":"publications","next":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","id":"/publications/2024-03-spatiallearning","collection":"publications","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Generative AI for visualization: State of the art and future directions. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"author\" content=\"Yilin Ye\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
298nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<meta property=\"og:description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yilin Ye\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yilin Ye\"},\"description\":\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
298nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\",\"headline\":\"Generative AI for visualization: State of the art and future directions.\",\"url\":\"https://zhanwang071.github.io//publications/2024-02-gen4ai\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
298 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<meta property=\"og:description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\",\"headline\":\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\",\"url\":\"https://zhanwang071.github.io//publications/2024-01-virtuwander\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
298 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"id":"/publications/2024-02-gen4ai","collection":"publications","next":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","excerpt":"","previous":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","excerpt":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
298nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","previous":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","id":"/publications/2024-01-virtuwander","collection":"publications","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"id":"/publications/2024-02-gen4ai","collection":"publications","next":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","id":"/publications/2024-03-spatiallearning","collection":"publications","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Generative AI for visualization: State of the art and future directions. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"author\" content=\"Yilin Ye\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
298nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<meta property=\"og:description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yilin Ye\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yilin Ye\"},\"description\":\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
298nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\",\"headline\":\"Generative AI for visualization: State of the art and future directions.\",\"url\":\"https://zhanwang071.github.io//publications/2024-02-gen4ai\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
298 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"id":"/publications/2024-03-spatiallearning","collection":"publications","next":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","excerpt":"","previous":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","id":"/publications/2024-03-spatiallearning","collection":"publications","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"id":"/publications/2024-04-isea","collection":"publications","next":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","id":"/publications/2025-01-cinefolio","collection":"publications","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["
298Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Staying vigilant in the Age of AI: From content generation to content authentication. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"author\" content=\"Yufan Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yufan Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yufan Li\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Staying vigilant in the Age of AI: From content generation to content authentication.\",\"url\":\"https://zhanwang071.github.io//publications/2024-04-isea\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
298 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"author\" content=\"Yifei Zhang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yifei Zhang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yifei Zhang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\",\"url\":\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
298 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Generative AI for visualization: State of the art and future directions. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"author\" content=\"Yilin Ye\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By
298summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<meta property=\"og:description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yilin Ye\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yilin Ye\"},\"description\":\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
298nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\",\"headline\":\"Generative AI for visualization: State of the art and future directions.\",\"url\":\"https://zhanwang071.github.io//publications/2024-02-gen4ai\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
298 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"}"> 299 <h3 id="/publications/2024-02-gen4ai">Generative AI for visualization: State of the art and future directions.</h3> 300 <div class="authors"> 301 Yilin Ye, Jianing Hao, Yihan Hou, <b>Zhan Wang</b>, Shishi Xiao, Yuyu Luo, Wei Zeng. 302 </div> 303 304 <div class="venue"> 305 306 Visual Informatics<!-- 307 308 --> 309 2024 310 </div> 311 312 313 314 315 316 317 318 319 320 321 <div class="extra-links"> 322 323 <a href="http://doi.org/10.1016/j.visinf.2024.04.003"> 324 <i class="fas fa-anchor" aria-hidden="true"></i> DOI 325 </a> 326 327 328 <a href="/papers/2024-gen4ai.pdf"> 329 <i class="far fa-file-pdf" aria-hidden="true"></i> PDF 330 </a> 331 332 333 <!-- --> 334 335 336 337 338 339 340 <!-- --> 341 342 343 </div> 344 345</div> 346 347<div class="publication" data-pub="{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","excerpt":"","previous":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","excerpt":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
347nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","previous":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","excerpt":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","previous":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","id":"/publications/2023-02-storymap","collection":"publications","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"id":"/publications/2024-01-virtuwander","collection":"publications","next":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","id":"/publications/2024-02-gen4ai","collection":"publications","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<meta property=\"og:description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\",\"headline\":\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\",\"url\":\"https://zhanwang071.github.io//publications/2024-01-virtuwander\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
347 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"id":"/publications/2024-02-gen4ai","collection":"publications","next":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","excerpt":"","previous":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","id":"/publications/2024-02-gen4ai","collection":"publications","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"id":"/publications/2024-03-spatiallearning","collection":"publications","next":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","id":"/publications/2024-04-isea","collection":"publications","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"author\" content=\"Yifei Zhang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yifei Zhang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yifei Zhang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\",\"url\":\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
347 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Generative AI for visualization: State of the art and future directions. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"author\" content=\"Yilin Ye\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By
347summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<meta property=\"og:description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yilin Ye\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yilin Ye\"},\"description\":\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
347nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\",\"headline\":\"Generative AI for visualization: State of the art and future directions.\",\"url\":\"https://zhanwang071.github.io//publications/2024-02-gen4ai\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
347 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"id":"/publications/2024-03-spatiallearning","collection":"publications","next":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","excerpt":"","previous":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","excerpt":"","previous":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","id":"/publications/2024-02-gen4ai","collection":"publications","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["
347Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"id":"/publications/2024-03-spatiallearning","collection":"publications","next":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","id":"/publications/2024-04-isea","collection":"publications","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"author\" content=\"Yifei Zhang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yifei Zhang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yifei Zhang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\",\"url\":\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
347 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"id":"/publications/2024-04-isea","collection":"publications","next":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","excerpt":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","previous":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","id":"/publications/2024-04-isea","collection":"publications","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"id":"/publications/2025-01-cinefolio","collection":"publications","next":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","id":"/publications/2025-02-vizta","collection":"publications","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["
347Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<meta property=\"og:description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
347sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\",\"headline\":\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\",\"url\":\"https://zhanwang071.github.io//publications/2025-01-cinefolio\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
347sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["
347Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Staying vigilant in the Age of AI: From content generation to content authentication. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"author\" content=\"Yufan Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yufan Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yufan Li\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Staying vigilant in the Age of AI: From content generation to content authentication.\",\"url\":\"https://zhanwang071.github.io//publications/2024-04-isea\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
347 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"author\" content=\"Yifei Zhang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yifei Zhang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yifei Zhang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\",\"url\":\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
347 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"}"> 348 <h3 id="/publications/2024-03-spatiallearning">Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives</h3> 349 <div class="authors"> 350 Yifei Zhang, Lizhe Chen, Jun-Hsiang Yao, <b>Zhan Wang</b>, Siming Chen. 351 </div> 352 353 <div class="venue"> 354 355 <a href="https://chinavis.org/2024/"> 356 357 The China Visualization and Visual Analytics Conference<!-- 358 359 --> in Hong Kong, China<!-- 360 361 --></a> 362 2024 363 </div> 364 365 366 367 368 369 370 371 372 373 374 <div class="extra-links"> 375 376 377 <a href="/papers/2024-spatiallearning.pdf"> 378 <i class="far fa-file-pdf" aria-hidden="true"></i> PDF 379 </a> 380 381 382 <!-- --> 383 384 385 386 <a href="/papers/2024-spatiallearning.mp4"> 387 <i class="fas fa-film" aria-hidden="true"></i> Video 388 </a> 389 390 391 392 393 <!-- --> 394 395 396 </div> 397 398</div> 399 400<div class="publication" data-pub="{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","excerpt":"","previous":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","excerpt":"","previous":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","excerpt":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
400nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","previous":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","id":"/publications/2024-01-virtuwander","collection":"publications","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"id":"/publications/2024-02-gen4ai","collection":"publications","next":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","id":"/publications/2024-03-spatiallearning","collection":"publications","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Generative AI for visualization: State of the art and future directions. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"author\" content=\"Yilin Ye\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
400nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<meta property=\"og:description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yilin Ye\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yilin Ye\"},\"description\":\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
400nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\",\"headline\":\"Generative AI for visualization: State of the art and future directions.\",\"url\":\"https://zhanwang071.github.io//publications/2024-02-gen4ai\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
400 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"id":"/publications/2024-03-spatiallearning","collection":"publications","next":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","excerpt":"","previous":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","id":"/publications/2024-03-spatiallearning","collection":"publications","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"id":"/publications/2024-04-isea","collection":"publications","next":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","id":"/publications/2025-01-cinefolio","collection":"publications","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["
400Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Staying vigilant in the Age of AI: From content generation to content authentication. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"author\" content=\"Yufan Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yufan Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yufan Li\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Staying vigilant in the Age of AI: From content generation to content authentication.\",\"url\":\"https://zhanwang071.github.io//publications/2024-04-isea\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
400 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"author\" content=\"Yifei Zhang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yifei Zhang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yifei Zhang\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives\",\"url\":\"https://zhanwang071.github.io//publications/2024-03-spatiallearning\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
400 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"id":"/publications/2024-04-isea","collection":"publications","next":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","excerpt":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","previous":{"path":"_publications/2024-04-isea.html","relative_path":"_publications/2024-04-isea.html","excerpt":"","previous":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","id":"/publications/2024-03-spatiallearning","collection":"publications","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"id":"/publications/2024-04-isea","collection":"publications","next":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","id":"/publications/2025-01-cinefolio","collection":"publications","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["
400Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Staying vigilant in the Age of AI: From content generation to content authentication. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"author\" content=\"Yufan Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yufan Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yufan Li\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Staying vigilant in the Age of AI: From content generation to content authentication.\",\"url\":\"https://zhanwang071.github.io//publications/2024-04-isea\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
400 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"},"id":"/publications/2025-01-cinefolio","collection":"publications","next":{"path":"_publications/2025-02-vizta.html","relative_path":"_publications/2025-02-vizta.html","excerpt":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","previous":{"path":"_publications/2025-01-cinefolio.html","relative_path":"_publications/2025-01-cinefolio.html","id":"/publications/2025-01-cinefolio","collection":"publications","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"id":"/publications/2025-02-vizta","collection":"publications","next":{"path":"_publications/2025-03-drama.html","relative_path":"_publications/2025-03-drama.html","id":"/publications/2025-03-drama","collection":"publications","url":"/publications/2025-03-drama","draft":false,"categories":[],"layout":"publication","year":2025,"title":"Centennial Drama Reimagined: An Immersive Experience of Intangible Cultural Heritage through Contextual Storytelling in Virtual Reality.","authors":["Jian Yu","Zhan Wang","Yifan Cao","Hao Cui","Wei Zeng"],"type":["
400Journal"],"venue":"Journal on Computing and Cultural Heritage","venue_tags":["JOCCH"],"tags":["Cultural Heritage","Storytelling","Virtual Reality"],"short_doi":"10.1145/3705613","highlight":false,"slug":"2025-03-drama","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<meta property=\"og:description\" content=\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-02-vizta\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\",\"headline\":\"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.\",\"url\":\"https://zhanwang071.github.io//publications/2025-02-vizta\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
400 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Comprehending visualizations requires readers to interpret visual encoding and the underlying meanings actively. This poses challenges for visualization novices, particularly when interpreting distributional visualizations that depict statistical uncertainty. Advancements in LLM-based conversational interfaces show promise in promoting visualization comprehension. However, they fail to provide contextual explanations at fine-grained granularity, and chart readers are still required to mentally bridge visual information and textual explanations during conversations. Our formative study highlights the expectations for both lexical and visual feedback, as well as the importance of explicitly linking these two modalities throughout the conversation. The findings motivate the design of VizTA, a visualization teaching assistant that leverages the fusion of visual and lexical feedback to help readers better comprehend visualization. VizTA features a semantic-aware conversational agent capable of explaining contextual information within visualizations and employs a visual-lexical fusion design to facilitate chart-centered conversation. A between-subject study with 24 participants demonstrates the effectiveness of VizTA in supporting the understanding and reasoning tasks of distributional visualization across multiple scenarios.","url":"/publications/2025-02-vizta","draft":false,"categories":[],"layout":"publication","year":2025,"title":"VizTA: Enhancing Comprehension of Distributional Visualization with Visual0-Lexical Fused Conversational Interface.","authors":["Liangwei Wang","Zhan Wang","Shishi Xiao","Le Liu","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Computer Graphics Forum","venue_tags":["CGF"],"tags":["Storytelling","Integration of Visualization and Text"],"short_doi":"10.1111/cgf.70110","highlight":false,"slug":"2025-02-vizta","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
400sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<meta property=\"og:description\" content=\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2025-01-cinefolio\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immer
400sive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\",\"headline\":\"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.\",\"url\":\"https://zhanwang071.github.io//publications/2025-01-cinefolio\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
400 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Narrative visualization facilitates data presentation and communicates insights, while virtual reality can further enhance immersive and engaging experiences. The combination of these two research interests shows the potential to revolutionize the way data is presented and understood. Within the realm of narrative visualization, empirical evidence has particularly highlighted the importance of camera planning. However, existing works primarily rely on user-intensive manipulation of the camera, with little effort put into automating the process. To fill the gap, this paper proposes CineFolio, a semi-automated camera planning method to reduce manual effort and enhance user experience in immersive narrative visualization. CineFolio combines cinematic theories with graphics criteria, considering both information delivery and aesthetic enjoyment to ensure a comfortable and engaging experience. Specifically, we parametrize the considerations into optimizable camera properties and solve it as a constraint satisfaction problem (CSP) to realize common camera types for narrative visualization, namely overview camera for absorbing the scale, focus camera for detailed views, moving camera for animated transitions, and user-controlled camera allowing users to provide inputs to camera planning. We demonstrate the feasibility of our approach with cases of various data and chart types. To further evaluate our approach, we conducted a within-subject user study, comparing our automated method with manual camera control, and the results confirm both effectiveness of the guided navigation and expressiveness of the cinematic design for narrative visualization.","url":"/publications/2025-01-cinefolio","draft":false,"categories":[],"layout":"publication","year":2025,"title":"CineFolio: Cinematography-Guided Camera Planning for Immersive Narrative Visualization.","authors":["Zhan Wang","Qian Zhu","David Yip","Fugee Tsung","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Virtual Reality","Narrative Visualization","Camera Planning","Cinematography"],"short_doi":"10.1016/j.visinf.2025.100259","highlight":true,"pdf":"/papers/2025-cinefolio.pdf","slug":"2025-01-cinefolio","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Staying vigilant in the Age of AI: From content generation to content authentication. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"author\" content=\"Yufan Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<meta property=\"og:description\" content=\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-04-isea\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Staying vigilant in the Age of AI: From content generation to content authentication.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yufan Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yufan Li\"},\"description\":\"The PhD at the Hong Kong University of Science and Technology (Guangzhou). My research interests are data storytelling, virtual reality, and human-computer interaction. BSc from Fudan University.\",\"headline\":\"Staying vigilant in the Age of AI: From content generation to content authentication.\",\"url\":\"https://zhanwang071.github.io//publications/2024-04-isea\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n \n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
400 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"","url":"/publications/2024-04-isea","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Staying vigilant in the Age of AI: From content generation to content authentication.","authors":["Yufan Li","Zhan Wang","Theo Papatheodorou"],"type":["Conference"],"venue":"International Symposium on Electronic Art","venue_location":"Brisbane, Australia","venue_tags":["ISEA"],"tags":["Large Language Model","Exploratory Studies"],"venue_url":"https://isea2024.isea-international.org/welcome/","highlight":false,"pdf":"/papers/2024-fakenews.pdf","link":"https://www.cacr-symp.com/","code":"https://github.com/ZhanWang071/YOAS2023","video":"https://vimeo.com/831242498?share=copy","slug":"2024-04-isea","ext":".html"}"> 401 <h3 id="/publications/2024-04-isea">Staying vigilant in the Age of AI: From content generation to content authentication.</h3> 402 <div class="authors"> 403 Yufan Li, <b>Zhan Wang</b>, Theo Papatheodorou. 404 </div> 405 406 <div class="venue"> 407 408 <a href="https://isea2024.isea-international.org/welcome/"> 409 410 International Symposium on Electronic Art<!-- 411 412 --> in Brisbane, Australia<!-- 413 414 --></a> 415 2024 416 </div> 417 418 419 420 421 422 423 424 425 426 427 <div class="extra-links"> 428 429 430 <a href="/papers/2024-fakenews.pdf"> 431 <i class="far fa-file-pdf" aria-hidden="true"></i> PDF 432 </a> 433 434 435 <a href="https://www.cacr-symp.com/"> 436 <i class="fas fa-link" aria-hidden="true"></i> Project 437 </a> 438 439 <!-- --> 440 441 442 443 <a href="https://vimeo.com/831242498?share=copy"> 444 <i class="fas fa-film" aria-hidden="true"></i> Video 445 </a> 446 447 448 449 450 <!-- --> 451 452 453 </div> 454 455</div> 456 457<h2 id="y2023" class="year">2023</h2> 458 459<div class="publication" data-pub="{"path":"_publications/2023-01-geocamera.html","relative_path":"_publications/2023-01-geocamera.html","excerpt":"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.","previous":null,"id":"/publications/2023-01-geocamera","collection":"publications","next":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","excerpt":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","previous":{"path":"_publications/2023-01-geocamera.html","relative_path":"_publications/2023-01-geocamera.html","excerpt":"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.","previous":null,"id":"/publications/2023-01-geocamera","collection":"publications","next":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","id":"/publications/2023-02-storymap","collection":"publications","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["
459Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\" />\n<meta name=\"author\" content=\"Wenchao Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\" />\n<meta property=\"og:description\" content=\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-01-geocamera\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-01-geocamera\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Wenchao Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Wenchao Li\"},\"description\":\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\",\"headline\":\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\",\"url\":\"https://zhanwang071.github.io//publications/2023-01-geocamera\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
459 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.","url":"/publications/2023-01-geocamera","draft":false,"categories":[],"layout":"publication","year":2023,"title":"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.","authors":["Wenchao Li","Zhan Wang","Yun Wang","Di Weng","Liwenhan Xie","Siming Chen","Haidong Zhang","Huamin Qu"],"type":["Conference"],"venue":"Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems","venue_location":"Hamburg, Germany","venue_tags":["CHI"],"tags":["Human-centered computing","Interactive systems and tools"],"venue_url":"https://chi2023.acm.org/","arxiv":"2303.06460","short_doi":"10.1145/3544548.3581470","highlight":true,"pdf":"https://dl.acm.org/doi/pdf/10.1145/3544548.3581470","video":"https://www.youtube.com/watch?v=SFBmBSHjKpA","slug":"2023-01-geocamera","ext":".html"},"id":"/publications/2023-02-storymap","collection":"publications","next":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","excerpt":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","previous":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","id":"/publications/2023-02-storymap","collection":"publications","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["
459Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"id":"/publications/2024-01-virtuwander","collection":"publications","next":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","id":"/publications/2024-02-gen4ai","collection":"publications","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<meta property=\"og:description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\",\"headline\":\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\",\"url\":\"https://zhanwang071.github.io//publications/2024-01-virtuwander\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
459 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<meta property=\"og:description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\",\"headline\":\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\",\"url\":\"https://zhanwang071.github.io//publications/2023-02-storymap\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
459 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\" />\n<meta name=\"author\" content=\"Wenchao Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\" />\n<meta property=\"og:description\" content=\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-01-geocamera\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-01-geocamera\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Wenchao Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Wenchao Li\"},\"description\":\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\",\"headline\":\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\",\"url\":\"https://zhanwang071.github.io//publications/2023-01-geocamera\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
459 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.","url":"/publications/2023-01-geocamera","draft":false,"categories":[],"layout":"publication","year":2023,"title":"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.","authors":["Wenchao Li","Zhan Wang","Yun Wang","Di Weng","Liwenhan Xie","Siming Chen","Haidong Zhang","Huamin Qu"],"type":["Conference"],"venue":"Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems","venue_location":"Hamburg, Germany","venue_tags":["CHI"],"tags":["Human-centered computing","Interactive systems and tools"],"venue_url":"https://chi2023.acm.org/","arxiv":"2303.06460","short_doi":"10.1145/3544548.3581470","highlight":true,"pdf":"https://dl.acm.org/doi/pdf/10.1145/3544548.3581470","video":"https://www.youtube.com/watch?v=SFBmBSHjKpA","slug":"2023-01-geocamera","ext":".html"}"> 460 <h3 id="/publications/2023-01-geocamera">GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.</h3> 461 <div class="authors"> 462 Wenchao Li, <b>Zhan Wang</b>, Yun Wang, Di Weng, Liwenhan Xie, Siming Chen, Haidong Zhang, Huamin Qu. 463 </div> 464 465 <div class="venue"> 466 467 <a href="https://chi2023.acm.org/"> 468 469 Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems<!-- 470 471 --> in Hamburg, Germany<!-- 472 473 --></a> 474 2023 475 </div> 476 477 478 479 480 481 482 483 484 485 486 <div class="extra-links"> 487 488 <a href="http://doi.org/10.1145/3544548.3581470"> 489 <i class="fas fa-anchor" aria-hidden="true"></i> DOI 490 </a> 491 492 493 <a href="https://dl.acm.org/doi/pdf/10.1145/3544548.3581470"> 494 <i class="far fa-file-pdf" aria-hidden="true"></i> PDF 495 </a> 496 497 498 <!-- 499 <a href="https://ar5iv.org/html/2303.06460"> 500 <i class="fab fa-html5" aria-hidden="true"></i> arXiv 501 </a> 502 --> 503 504 505 <a href="https://arxiv.org/abs/2303.06460"> 506 <i class="fas fa-archive" aria-hidden="true"></i> arXiv 507 </a> 508 509 510 <a href="https://www.youtube.com/watch?v=SFBmBSHjKpA"> 511 <i class="fas fa-film" aria-hidden="true"></i> Video 512 </a> 513 514 515 516 517 <!-- 518 <a href="https://arxiv2bibtex.org/?q=2303.06460"> 519 <i class="fas fa-book" aria-hidden="true"></i> Bibtex 520 </a> 521 --> 522 523 524 </div> 525 526</div> 527 528<div class="publication" data-pub="{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","excerpt":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","previous":{"path":"_publications/2023-01-geocamera.html","relative_path":"_publications/2023-01-geocamera.html","excerpt":"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.","previous":null,"id":"/publications/2023-01-geocamera","collection":"publications","next":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","excerpt":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","previous":{"path":"_publications/2023-01-geocamera.html","relative_path":"_publications/2023-01-geocamera.html","id":"/publications/2023-01-geocamera","collection":"publications","url":"/publications/2023-01-geocamera","draft":false,"categories":[],"layout":"publication","year":2023,"title":"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.","authors":["Wenchao Li","Zhan Wang","Yun Wang","Di Weng","Liwenhan Xie","Siming Chen","Haidong Zhang","Huamin Qu"],"type":["Conference"],"venue":"Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems","venue_location":"Hamburg, Germany","venue_tags":["CHI"],"tags":["Human-centered computing","Interactive systems and tools"],"venue_url":"https://chi2023.acm.org/","arxiv":"2303.06460","short_doi":"10.1145/3544548.3581470","highlight":true,"pdf":"https://dl.acm.org/doi/pdf/10.1145/3544548.3581470","video":"https://www.youtube.com/watch?v=SFBmBSHjKpA","slug":"2023-01-geocamera","ext":".html"},"id":"/publications/2023-02-storymap","collection":"publications","next":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","id":"/publications/2024-01-virtuwander","collection":"publications","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<meta property=\"og:description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with grap
528hic-based approach.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\",\"headline\":\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\",\"url\":\"https://zhanwang071.github.io//publications/2023-02-storymap\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
528 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\" />\n<meta name=\"author\" content=\"Wenchao Li\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\" />\n<meta property=\"og:description\" content=\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-01-geocamera\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-01-geocamera\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Wenchao Li\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Wenchao Li\"},\"description\":\"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\",\"headline\":\"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.\",\"url\":\"https://zhanwang071.github.io//publications/2023-01-geocamera\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
528 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"In geographic data videos, camera movements are frequently used and combined to present information from multiple perspectives. However, creating and editing camera movements requires significant time and professional skills. This work aims to lower the barrier of crafting diverse camera movements for geographic data videos. First, we analyze a corpus of 66 geographic data videos and derive a design space of camera movements with a dimension for geospatial targets and one for narrative purposes. Based on the design space, we propose a set of adaptive camera shots and further develop an interactive tool called GeoCamera. This interactive tool allows users to flexibly design camera movements for geographic visualizations. We verify the expressiveness of our tool through case studies and evaluate its usability with a user study. The participants find that the tool facilitates the design of camera movements.","url":"/publications/2023-01-geocamera","draft":false,"categories":[],"layout":"publication","year":2023,"title":"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.","authors":["Wenchao Li","Zhan Wang","Yun Wang","Di Weng","Liwenhan Xie","Siming Chen","Haidong Zhang","Huamin Qu"],"type":["Conference"],"venue":"Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems","venue_location":"Hamburg, Germany","venue_tags":["CHI"],"tags":["Human-centered computing","Interactive systems and tools"],"venue_url":"https://chi2023.acm.org/","arxiv":"2303.06460","short_doi":"10.1145/3544548.3581470","highlight":true,"pdf":"https://dl.acm.org/doi/pdf/10.1145/3544548.3581470","video":"https://www.youtube.com/watch?v=SFBmBSHjKpA","slug":"2023-01-geocamera","ext":".html"},"id":"/publications/2023-02-storymap","collection":"publications","next":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","excerpt":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","previous":{"path":"_publications/2023-02-storymap.html","relative_path":"_publications/2023-02-storymap.html","excerpt":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","previous":{"path":"_publications/2023-01-geocamera.html","relative_path":"_publications/2023-01-geocamera.html","id":"/publications/2023-01-geocamera","collection":"publications","url":"/publications/2023-01-geocamera","draft":false,"categories":[],"layout":"publication","year":2023,"title":"GeoCamera: Telling Stories in Geographic Visualizations with Camera Movements.","authors":["Wenchao Li","Zhan Wang","Yun Wang","Di Weng","Liwenhan Xie","Siming Chen","Haidong Zhang","Huamin Qu"],"type":["Conference"],"venue":"Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems","venue_location":"Hamburg, Germany","venue_tags":["CHI"],"tags":["Human-centered computing","Interactive systems and tools"],"venue_url":"https://chi2023.acm.org/","arxiv":"2303.06460","short_doi":"10.1145/3544548.3581470","highlight":true,"pdf":"https://dl.acm.org/doi/pdf/10.1145/3544548.3581470","video":"https://www.youtube.com/watch?v=SFBmBSHjKpA","slug":"2023-01-geocamera","ext":".html"},"id":"/publications/2023-02-storymap","collection":"publications","next":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","id":"/publications/2024-01-virtuwander","collection":"publications","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<meta property=\"og:description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\",\"headline\":\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\",\"url\":\"https://zhanwang071.github.io//publications/2023-02-storymap\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\&quo
528t; aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length; i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["
528Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"},"id":"/publications/2024-01-virtuwander","collection":"publications","next":{"path":"_publications/2024-02-gen4ai.html","relative_path":"_publications/2024-02-gen4ai.html","excerpt":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","previous":{"path":"_publications/2024-01-virtuwander.html","relative_path":"_publications/2024-01-virtuwander.html","id":"/publications/2024-01-virtuwander","collection":"publications","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"id":"/publications/2024-02-gen4ai","collection":"publications","next":{"path":"_publications/2024-03-spatiallearning.html","relative_path":"_publications/2024-03-spatiallearning.html","id":"/publications/2024-03-spatiallearning","collection":"publications","url":"/publications/2024-03-spatiallearning","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Enhancing Spatial Learning through Visualization Cues: A Comparative Study of Virtuality and Perspectives","authors":["Yifei Zhang","Lizhe Chen","Jun-Hsiang Yao","Zhan Wang","Siming Chen"],"type":["Conference"],"venue":"The China Visualization and Visual Analytics Conference","venue_location":"Hong Kong, China","venue_tags":["ChinaVis"],"tags":["Empirical Studies","Virtual reality"],"venue_url":"https://chinavis.org/2024/","pdf":"/papers/2024-spatiallearning.pdf","video":"/papers/2024-spatiallearning.mp4","slug":"2024-03-spatiallearning","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Generative AI for visualization: State of the art and future directions. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"author\" content=\"Yilin Ye\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
528nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<meta property=\"og:description\" content=\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-02-gen4ai\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Generative AI for visualization: State of the art and future directions.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Yilin Ye\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Yilin Ye\"},\"description\":\"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-e
528nd GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\",\"headline\":\"Generative AI for visualization: State of the art and future directions.\",\"url\":\"https://zhanwang071.github.io//publications/2024-02-gen4ai\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
528 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Generative AI (GenAI) has witnessed remarkable progress in recent years and demonstrated impressive performance in various generation tasks in different domains such as computer vision and computational design. Many researchers have attempted to integrate GenAI into visualization framework, leveraging the superior generative capacity for different operations. Concurrently, recent major breakthroughs in GenAI like diffusion models and large language models have also drastically increased the potential of GenAI4VIS. From a technical perspective, this paper looks back on previous visualization studies leveraging GenAI and discusses the challenges and opportunities for future research. Specifically, we cover the applications of different types of GenAI methods including sequence, tabular, spatial and graph generation techniques for different tasks of visualization which we summarize into four major stages: data enhancement, visual mapping generation, stylization and interaction. For each specific visualization sub-task, we illustrate the typical data and concrete GenAI algorithms, aiming to provide in-depth understanding of the state-of-the-art GenAI4VIS techniques and their limitations. Furthermore, based on the survey, we discuss three major aspects of challenges and research opportunities including evaluation, dataset, and the gap between end-to-end GenAI methods and visualizations. By summarizing different generation algorithms, their current applications and limitations, this paper endeavors to provide useful insights for future GenAI4VIS research.","url":"/publications/2024-02-gen4ai","draft":false,"categories":[],"layout":"publication","year":2024,"title":"Generative AI for visualization: State of the art and future directions.","authors":["Yilin Ye","Jianing Hao","Yihan Hou","Zhan Wang","Shishi Xiao","Yuyu Luo","Wei Zeng"],"type":["Journal"],"venue":"Visual Informatics","venue_tags":["VI"],"tags":["Survey","Generative AI","Data Visualization"],"short_doi":"10.1016/j.visinf.2024.04.003","highlight":true,"pdf":"/papers/2024-gen4ai.pdf","slug":"2024-02-gen4ai","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"author\" content=\"Zhan Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<meta property=\"og:description\" content=\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2024-01-virtuwander\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Zhan Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Zhan Wang\"},\"description\":\"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\",\"headline\":\"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.\",\"url\":\"https://zhanwang071.github.io//publications/2024-01-virtuwander\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
528 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Tour guidance in virtual museums encourages multi-modal interactions to boost user experiences, concerning engagement, immersion, and spatial awareness. Nevertheless, achieving the goal is challenging due to the complexity of comprehending diverse user needs and accommodating personalized user preferences. Informed by a formative study that characterizes guidance-seeking contexts, we establish a multi-modal interaction design framework for virtual tour guidance. We then design VirtuWander, a two-stage innovative system using domain-oriented large language models to transform user inquiries into diverse guidance-seeking contexts and facilitate multi-modal interactions. The feasibility and versatility of VirtuWander are demonstrated with virtual guiding examples that encompass various touring scenarios and cater to personalized preferences. We further evaluate VirtuWander through a user study within an immersive simulated museum. The results suggest that our system enhances engaging virtual tour experiences through personalized communication and knowledgeable assistance, indicating its potential for expanding into real-world scenarios.","url":"/publications/2024-01-virtuwander","draft":false,"categories":[],"layout":"publication","year":2024,"title":"VirtuWander: Enhancing Multi-modal Interaction for Virtual Tour Guidance through Large Language Models.","authors":["Zhan Wang","Linping Yuan","Liangwei Wang","Bingchuan Jiang","Wei Zeng"],"type":["Conference"],"venue":"Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems","venue_location":"Hawaii, US","venue_tags":["CHI"],"tags":["Human-centered computing","Virtual reality"],"venue_url":"https://chi2024.acm.org/","arxiv":"2401.11923","short_doi":"10.1145/3613904.3642235","highlight":true,"pdf":"https://arxiv.org/pdf/2401.11923.pdf","video":"https://youtu.be/Dt3g4NMFmUc","slug":"2024-01-virtuwander","ext":".html"},"output":"<!DOCTYPE html>\n<html lang=\"en-US\">\n <head>\n <meta charset=\"utf-8\">\n <meta http-equiv=\"x-ua-compatible\" content=\"ie=edge\">\n <meta name=\"viewport\" content=\"width=device-width, initial-scale=1\">\n <meta name=\"application-name\" content=\"Zhan Wang\">\n <meta name=\"theme-color\" content=\"#fba211\">\n <link rel=\"shortcut icon\" href=\"/favicon.png\"/>\n <link rel=\"icon\" type=\"image/png\" href=\"/favicon.png\" sizes=\"250x250\" />\n\n <!-- Begin Jekyll SEO tag v2.8.0 -->\n<title>Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach. | Zhan Wang</title>\n<meta name=\"generator\" content=\"Jekyll v3.10.0\" />\n<meta property=\"og:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"author\" content=\"Liangwei Wang\" />\n<meta property=\"og:locale\" content=\"en_US\" />\n<meta name=\"description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<meta property=\"og:description\" content=\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\" />\n<link rel=\"canonical\" href=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:url\" content=\"https://zhanwang071.github.io//publications/2023-02-storymap\" />\n<meta property=\"og:site_name\" content=\"Zhan Wang\" />\n<meta property=\"og:type\" content=\"website\" />\n<meta name=\"twitter:card\" content=\"summary\" />\n<meta property=\"twitter:title\" content=\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\" />\n<meta name=\"twitter:site\" content=\"@zoeywuwa\" />\n<meta name=\"twitter:creator\" content=\"@Liangwei Wang\" />\n<script type=\"application/ld+json\">\n{\"@context\":\"https://schema.org\",\"@type\":\"WebPage\",\"author\":{\"@type\":\"Person\",\"name\":\"Liangwei Wang\"},\"description\":\"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\",\"headline\":\"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.\",\"url\":\"https://zhanwang071.github.io//publications/2023-02-storymap\"}</script>\n<!-- End Jekyll SEO tag -->\n\n\n <link rel=\"alternate\" type=\"application/rss+xml\" title=\"Zhan Wang\" href=\"https://zhanwang071.github.io//feed.xml\">\n\n <link href=\"https://use.fontawesome.com/releases/v6.2.1/css/all.css\" rel=\"stylesheet\">\n <link href=\"/styles.css\" rel=\"stylesheet\">\n</head>\n\n <body>\n \n<header class=\"page-header\">\n <nav class=\"container\">\n <a class=\"site-title\" href=\"/\">Zhan Wang</a>\n\n <a href=\"/projects/\" >Projects</a>\n <a href=\"/publications/\" aria-current=\"page\">Publications</a>\n <!-- <a href=\"/talks/\" >Talks</a> -->\n <!-- <a href=\"/blog/\" >Blog</a> -->\n <a href=\"/cv/\" >CV</a>\n\n <span class=\"external\">\n <!-- <a rel=\"me\" href=\"https://vis.social/@dom\"><i class=\"fab fa-mastodon\" aria-hidden=\"true\"></i> Mastodon</a> -->\n <a href=\"https://scholar.google.com/citations?user=TRyWRYUAAAAJ&hl=en\"><i class=\"fab fa-google\" aria-hidden=\"true\"></i> Scholar</a>\n <a href=\"https://github.com/zhanwang071\"><i class=\"fab fa-github\" aria-hidden=\"true\"></i> GitHub</a>\n </span>\n </nav>\n</header>\n\n\n <div class=\"page-content\">\n <section class=\"container \">\n Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n\n</section>\n\n </div>\n\n <footer>\n <div class=\"container\">\n <div class=\"footer-col\">\n Hello from Guangzhou.<br/>\n <abbr title=\"Last build on 2026-09-11\">September 2026</abbr>\n </div>\n <div class=\"footer-col site-desc\">\n The PhD from HKUST(GZ) working on immersive visualization. I'm <a href=\"https://twitter.com/zoeywuwa\">@zoeywuwa</a> on Twitter and <a rel=\"me\" href=\"https://www.instagram.com/zoeywang071/\">@zoeywang071</a> on Instagram.\n </div>\n <div class=\"footer-col\">template from <a href=\"https://github.com/domoritz/domoritz.github.io\">@domoritz</a></div>\n </div>\n</footer>\n\n <script>\n function trim(str) {\n return str.replace(/^\\s+|\\s+$/g, '');\n }\n var headers = document.querySelectorAll(\"h2, h3, h4, h5, h6\");\n for (var i=0; i<headers.length;
528 i++) {\n var h = headers[i];\n var name = h.getAttribute(\"id\");\n var title = h.innerHTML;\n h.innerHTML = '<a href=\"#' + name + '\" class=\"anchor\"><i class=\"fas fa-hashtag\"></i></a>' + trim(title);\n }\n</script>\n\n </body>\n</html>\n","content":"Story maps have been widely utilized for their ability to provide a visual and spatial framework for storytelling. However, existing story map tools have limitations in creating diverse narrative structures and providing interactive options, and they cannot effectively render maps for polar regions due to tile-based mapping constraints. In this paper, we propose a graphic-based approach to overcome these challenges and develop a workflow for creating story maps specifically designed for polar regions. Our primary goal is to provide heuristic strategies for story map design and explore the potential of story maps in visualizing and disseminating polar culture. We summarize the main design tasks involved in story map creation and introduce three map-based visual narrative strategies, i.e., attention cue, linkage between maps and charts, and cartographic interaction. Additionally, we delve into the importance of storyboard design, taking into account logic, time order, and map granularity. To demonstrate the effectiveness of our proposed story map design method, we create story map cases focusing on the exploration history of Antarctica. These cases showcase the diverse and interactive nature of the story maps created using our approach. We discuss the limitations and challenges in creating story maps and identify potential research opportunities from our study.\n","url":"/publications/2023-02-storymap","draft":false,"categories":[],"layout":"publication","year":2023,"title":"Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.","authors":["Liangwei Wang","Zhan Wang","Xi Zhao","Fugee Tsung","Wei Zeng"],"<!-- awards":["Best Paper Award -->"],"type":["Journal"],"venue":"VINCI'23 of The Visual Computer","<!-- venue_location":"Guangzhou, China -->","venue_tags":["VINCI"],"tags":["Visualization techniques","Visual storytelling"],"venue_url":"http://www.vinci-symp.org/","<!-- highlight":"false -->","pdf":"https://link.springer.com/article/10.1007/s00371-024-03489-x","slug":"2023-02-storymap","ext":".html"}"> 529 <h3 id="/publications/2023-02-storymap">Antarctica storytelling: creating interactive story maps for polar regions with graphic-based approach.</h3> 530 <div class="authors"> 531 Liangwei Wang, <b>Zhan Wang</b>, Xi Zhao, Fugee Tsung, Wei Zeng. 532 </div> 533 534 <div class="venue"> 535 536 <a href="http://www.vinci-symp.org/"> 537 538 VINCI'23 of The Visual Computer<!-- 539 540 --></a> 541 2023 542 </div> 543 544 545 546 547 548 549 550 551 552 553 <div class="extra-links"> 554 555 556 <a href="https://link.springer.com/article/10.1007/s00371-024-03489-x"> 557 <i class="far fa-file-pdf" aria-hidden="true"></i> PDF 558 </a> 559 560 561 <!-- --> 562 563 564 565 566 567 568 <!-- --> 569 570 571 </div> 572 573</div> 574 575<!--
575<script src="https://cdn.jsdelivr.net/npm/[email protected]/dist/itemsjs.min.js"></script>
575 -->
576<script> 577 (function(e){if("object"==typeof exports&&"undefined"!=typeof module)module.exports=e();else if("function"==typeof define&&define.amd)define([],e);else{var t;t="undefined"==typeof window?"undefined"==typeof global?"undefined"==typeof self?this:self:global:window,t.itemsjs=e()}})(function(){var e=Math.floor,T=Math.max,V=Math.min,G;return function(){function d(s,e,n){function t(l,r){if(!e[l]){if(!s[l]){var i="function"==typeof require&&require;if(!r&&i)return i(l,!0);if(o)return o(l,!0);var _=new Error("Cannot find module '"+l+"'");throw _.code="MODULE_NOT_FOUND",_}var a=e[l]={exports:{}};s[l][0].call(a.exports,function(e){var o=s[l][1][e];return t(o||e)},a,a.exports,d,s,e,n)}return e[l].exports}for(var o="function"==typeof require&&require,r=0;r<n.length;r++)t(n[r]);return t}return d}()({1:[function(e,t){"use strict";t.exports=e("./src/index")},{"./src/index":8}],2:[function(e,t){"use strict";function n(e){if(this.words=[],e)if(Symbol&&Symbol.iterator&&void 0!==e[Symbol.iterator]){const t=e[Symbol.iterator]();for(let e=t.next();!e.done;)this.add(e.value),e=t.next()}else for(let t=0;t<e.length;t++)this.add(e[t])}n.prototype.fromWords=function(e){const t=Object.create(n.prototype);return t.words=e,t},n.prototype.add=function(e){this.resize(e),this.words[e>>>5]|=1<<e},n.prototype.flip=function(e){this.resize(e),this.words[e>>>5]^=1<<e},n.prototype.clear=function(){this.words.length=0},n.prototype.remove=function(e){this.resize(e),this.words[e>>>5]&=~(1<<e)},n.prototype.isEmpty=function(){const e=this.words.length;for(let t=0;t<e;t++)if(0!==this.words[t])return!1;return!0},n.prototype.has=function(e){return 0!=(this.words[e>>>5]&1<<e)},n.prototype.checkedAdd=function(e){this.resize(e);const t=this.words[e>>>5],n=t|1<<e;return this.words[e>>>5]=n,(n^t)>>>e},n.prototype.trim=function(){let e=this.words.length;for(;0<e&&0===this.words[e-1];)e--;this.words=this.words.slice(0,e)},n.prototype.resize=function(e){for(let t=this.words.length;t<e+32>>>5;t++)this.words[t]=0},n.prototype.hammingWeight=function(e){return e-=1431655765&e>>>1,e=(858993459&e)+(858993459&e>>>2),16843009*(252645135&e+(e>>>4))>>>24},n.prototype.hammingWeight4=function(e,t,n,o){return e-=1431655765&e>>>1,t-=1431655765&t>>>1,n-=1431655765&n>>>1,o-=1431655765&o>>>1,e=(858993459&e)+(858993459&e>>>2),t=(858993459&t)+(858993459&t>>>2),n=(858993459&n)+(858993459&n>>>2),o=(858993459&o)+(858993459&o>>>2),e=252645135&e+(e>>>4),t=252645135&t+(t>>>4),n=252645135&n+(n>>>4),o=252645135&o+(o>>>4),16843009*(e+t+n+o)>>>24},n.prototype.size=function(){let e=0;const t=this.words.length,n=this.words;for(let o=0;o<t;o++)e+=this.hammingWeight(n[o]);return e},n.prototype.array=function(){const e=Array(this.size());let n=0;const t=this.words.length;for(let o=0,r;o<t;++o)for(r=this.words[o];0!=r;){const i=r&-r;e[n++]=(o<<5)+this.hammingWeight(0|i-1),r^=i}return e},n.prototype.forEach=function(e){const t=this.words.length;for(let n=0,o;n<t;++n)for(o=this.words[n];0!=o;){const r=o&-o;e((n<<5)+this.hammingWeight(0|r-1)),o^=r}},n.prototype[Symbol.iterator]=function*(){const e=this.words.length;for(let n=0,o;n<e;++n)for(o=this.words[n];0!=o;){const e=o&-o;yield(n<<5)+this.hammingWeight(0|e-1),o^=e}},n.prototype.clone=function(){const e=Object.create(n.prototype);return e.words=this.words.slice(),e},n.prototype.intersects=function(e){const t=V(this.words.length,e.words.length);for(let n=0;n<t;++n)if(0!=(this.words[n]&e.words[n]))return!0;return!1},n.prototype.intersection=function(e){const t=V(this.words.length,e.words.length);let n=0;for(;n+7<t;n+=8)this.words[n]&=e.words[n],this.words[n+1]&=e.words[n+1],this.words[n+2]&=e.words[n+2],this.words[n+3]&=e.words[n+3],this.words[n+4]&=e.words[n+4],this.words[n+5]&=e.words[n+5],this.words[n+6]&=e.words[n+6],this.words[n+7]&=e.words[n+7];for(;n<t;++n)this.words[n]&=e.words[n];const o=this.words.length;for(n=t;n<o;++n)this.words[n]=0;return this},n.prototype.intersection_size=function(e){const t=V(this.words.length,e.words.length);let n=0;for(let o=0;o<t;++o)n+=this.hammingWeight(this.words[o]&e.words[o]);return n},n.prototype.new_intersection=function(e){const t=Object.create(n.prototype),o=V(this.words.length,e.words.length);t.words=Array(o);let r=0;for(;r+7<o;r+=8)t.words[r]=this.words[r]&e.words[r],t.words[r+1]=this.words[r+1]&e.words[r+1],t.words[r+2]=this.words[r+2]&e.words[r+2],t.words[r+3]=this.words[r+3]&e.words[r+3],t.words[r+4]=this.words[r+4]&e.words[r+4],t.words[r+5]=this.words[r+5]&e.words[r+5],t.words[r+6]=this.words[r+6]&e.words[r+6],t.words[r+7]=this.words[r+7]&e.words[r+7];for(;r<o;++r)t.words[r]=this.words[r]&e.words[r];return t},n.prototype.equals=function(e){const t=V(this.words.length,e.words.length);for(let n=0;n<t;++n)if(this.words[n]!=e.words[n])return!1;if(this.words.length<e.words.length){const t=e.words.length;for(let n=this.words.length;n<t;++n)if(0!=e.words[n])return!1}else if(e.words.length<this.words.length){const t=this.words.length;for(let n=e.words.length;n<t;++n)if(0!=this.words[n])return!1}return!0},n.prototype.difference=function(e){const t=V(this.words.length,e.words.length);let n=0;for(;n+7<t;n+=8)this.words[n]&=~e.words[n],this.words[n+1]&=~e.words[n+1],this.words[n+2]&=~e.words[n+2],this.words[n+3]&=~e.words[n+3],this.words[n+4]&=~e.words[n+4],this.words[n+5]&=~e.words[n+5],this.words[n+6]&=~e.words[n+6],this.words[n+7]&=~e.words[n+7];for(;n<t;++n)this.words[n]&=~e.words[n];return this},n.prototype.difference2=function(e){const t=V(this.words.length,e.words.length);let n=0;for(;n+7<t;n+=8)e.words[n]=this.words[n]&~e.words[n],e.words[n+1]=this.words[n+1]&~e.words[n+1],e.words[n+2]=this.words[n+2]&~e.words[n+2],e.words[n+3]=this.words[n+3]&~e.words[n+3],e.words[n+4]=this.words[n+4]&~e.words[n+4],e.words[n+5]=this.words[n+5]&~e.words[n+5],e.words[n+6]=this.words[n+6]&~e.words[n+6],e.words[n+7]=this.words[n+7]&~e.words[n+7];for(;n<t;++n)e.words[n]=this.words[n]&~e.words[n];for(n=this.words.length-1;n>=t;--n)e.words[n]=this.words[n];return e.words=e.words.slice(0,this.words.length),e},n.prototype.new_difference=function(e){return this.clone().difference(e)},n.prototype.difference_size=function(e){const t=V(this.words.length,e.words.length);let n=0,o=0;for(;o<t;++o)n+=this.hammingWeight(this.words[o]&~e.words[o]);for(const t=this.words.length;o<t;++o)n+=this.hammingWeight(this.words[o]);return n},n.prototype.change=function(e){const t=V(this.words.length,e.words.length);let n=0;for(;n+7<t;n+=8)this.words[n]^=e.words[n],this.words[n+1]^=e.words[n+1],this.words[n+2]^=e.words[n+2],this.words[n+3]^=e.words[n+3],this.words[n+4]^=e.words[n+4],this.words[n+5]^=e.words[n+5],this.words[n+6]^=e.words[n+6],this.words[n+7]^=e.words[n+7];for(;n<t;++n)this.words[n]^=e.words[n];for(n=e.words.length-1;n>=t;--n)this.words[n]=e.words[n];return this},n.prototype.new_change=function(e){const t=Object.create(n.prototype),o=T(this.words.length,e.words.length);t.words=Array(o);const r=V(this.words.length,e.words.length);let i=0;for(;i+7<r;i+=8)t.words[i]=this.words[i]^e.words[i],t.words[i+1]=this.words[i+1]^e.words[i+1],t.words[i+2]=this.words[i+2]^e.words[i+2],t.words[i+3]=this.words[i+3]^e.words[i+3],t.words[i+4]=this.words[i+4]^e.words[i+4],t.w
577ords[i+5]=this.words[i+5]^e.words[i+5],t.words[i+6]=this.words[i+6]^e.words[i+6],t.words[i+7]=this.words[i+7]^e.words[i+7];for(;i<r;++i)t.words[i]=this.words[i]^e.words[i];const d=this.words.length;for(i=r;i<d;++i)t.words[i]=this.words[i];const a=e.words.length;for(i=r;i<a;++i)t.words[i]=e.words[i];return t},n.prototype.change_size=function(e){const t=V(this.words.length,e.words.length);let n=0,o=0;for(;o<t;++o)n+=this.hammingWeight(this.words[o]^e.words[o]);const r=this.words.length>e.words.length?this:e,i=r.words.length;for(;o<i;++o)n+=this.hammingWeight(r.words[o]);return n},n.prototype.toString=function(){return"{"+this.array().join(",")+"}"},n.prototype.union=function(e){const t=V(this.words.length,e.words.length);let n=0;for(;n+7<t;n+=8)this.words[n]|=e.words[n],this.words[n+1]|=e.words[n+1],this.words[n+2]|=e.words[n+2],this.words[n+3]|=e.words[n+3],this.words[n+4]|=e.words[n+4],this.words[n+5]|=e.words[n+5],this.words[n+6]|=e.words[n+6],this.words[n+7]|=e.words[n+7];for(;n<t;++n)this.words[n]|=e.words[n];if(this.words.length<e.words.length){this.resize((e.words.length<<5)-1);const n=e.words.length;for(let o=t;o<n;++o)this.words[o]=e.words[o]}return this},n.prototype.new_union=function(e){const t=Object.create(n.prototype),o=T(this.words.length,e.words.length);t.words=Array(o);const r=V(this.words.length,e.words.length);let i=0;for(;i+7<r;i+=8)t.words[i]=this.words[i]|e.words[i],t.words[i+1]=this.words[i+1]|e.words[i+1],t.words[i+2]=this.words[i+2]|e.words[i+2],t.words[i+3]=this.words[i+3]|e.words[i+3],t.words[i+4]=this.words[i+4]|e.words[i+4],t.words[i+5]=this.words[i+5]|e.words[i+5],t.words[i+6]=this.words[i+6]|e.words[i+6],t.words[i+7]=this.words[i+7]|e.words[i+7];for(;i<r;++i)t.words[i]=this.words[i]|e.words[i];const d=this.words.length;for(i=r;i<d;++i)t.words[i]=this.words[i];const a=e.words.length;for(i=r;i<a;++i)t.words[i]=e.words[i];return t},n.prototype.union_size=function(e){const t=V(this.words.length,e.words.length);let n=0;for(let o=0;o<t;++o)n+=this.hammingWeight(this.words[o]|e.words[o]);if(this.words.length<e.words.length){const t=e.words.length;for(let o=this.words.length;o<t;++o)n+=this.hammingWeight(0|e.words[o])}else{const t=this.words.length;for(let o=e.words.length;o<t;++o)n+=this.hammingWeight(0|this.words[o])}return n},t.exports=n},{}],3:[function(t,n,o){(function(t){(function(){(function(){function r(e,t,n){switch(n.length){case 0:return e.call(t);case 1:return e.call(t,n[0]);case 2:return e.call(t,n[0],n[1]);case 3:return e.call(t,n[0],n[1],n[2]);}return e.apply(t,n)}function i(e,t,n,o){for(var r=-1,i=null==e?0:e.length;++r<i;){var d=e[r];t(o,d,n(d),e)}return o}function d(e,t){for(var n=-1,o=null==e?0:e.length;++n<o&&!(!1===t(e[n],n,e)););return e}function a(e,t){for(var n=null==e?0:e.length;n--&&!(!1===t(e[n],n,e)););return e}function s(e,t){for(var n=-1,o=null==e?0:e.length;++n<o;)if(!t(e[n],n,e))return!1;return!0}function l(e,t){for(var n=-1,o=null==e?0:e.length,r=0,i=[];++n<o;){var d=e[n];t(d,n,e)&&(i[r++]=d)}return i}function p(e,t){var n=null==e?0:e.length;return!!n&&-1<x(e,t,0)}function c(e,t,n){for(var o=-1,r=null==e?0:e.length;++o<r;)if(n(t,e[o]))return!0;return!1}function u(e,t){for(var n=-1,o=null==e?0:e.length,r=Array(o);++n<o;)r[n]=t(e[n],n,e);return r}function h(e,t){for(var n=-1,o=t.length,r=e.length;++n<o;)e[r+n]=t[n];return e}function g(e,t,n,o){var r=-1,i=null==e?0:e.length;for(o&&i&&(n=e[++r]);++r<i;)n=t(n,e[r],r,e);return n}function f(e,t,n,o){var r=null==e?0:e.length;for(o&&r&&(n=e[--r]);r--;)n=t(n,e[r],r,e);return n}function y(e,t){for(var n=-1,o=null==e?0:e.length;++n<o;)if(t(e[n],n,e))return!0;return!1}function m(e){return e.split("")}function w(e){return e.match(St)||[]}function b(e,t,n){var o;return n(e,function(e,n,r){if(t(e,n,r))return o=n,!1}),o}function k(e,t,n,o){for(var r=e.length,i=n+(o?1:-1);o?i--:++i<r;)if(t(e[i],i,e))return i;return-1}function x(e,t,n){return t===t?X(e,t,n):k(e,v,n)}function j(e,t,n,o){for(var r=n-1,i=e.length;++r<i;)if(o(e[r],t))return r;return-1}function v(e){return e!==e}function S(e,t){var n=null==e?0:e.length;return n?E(e,t)/n:Ie}function A(e){return function(t){return null==t?void 0:t[e]}}function z(e){return function(t){return null==e?void 0:e[t]}}function I(e,t,n,o,r){return r(e,function(e,r,i){n=o?(o=!1,e):t(n,e,r,i)}),n}function O(e,t){var n=e.length;for(e.sort(t);n--;)e[n]=e[n].value;return e}function E(e,t){for(var n=-1,o=e.length,r;++n<o;){var i=t(e[n]);i!==void 0&&(r=r==void 0?i:r+i)}return r}function F(e,t){for(var n=-1,o=Array(e);++n<e;)o[n]=t(n);return o}function N(e,t){return u(t,function(t){return[t,e[t]]})}function R(e){return e?e.slice(0,oe(e)+1).replace(bt,""):e}function W(e){return function(t){return e(t)}}function C(e,t){return u(t,function(t){return e[t]})}function L(e,t){return e.has(t)}function P(e,t){for(var n=-1,o=e.length;++n<o&&-1<x(t,e[n],0););return n}function B(e,t){for(var n=e.length;n--&&-1<x(t,e[n],0););return n}function U(e,t){for(var n=e.length,o=0;n--;)e[n]===t&&++o;return o}function M(e){return"\\"+bn[e]}function D(e,t){return null==e?void 0:e[t]}function q(e){return hn.test(e)}function $(e){return gn.test(e)}function J(e){for(var t=[],n;!(n=e.next()).done;)t.push(n.value);return t}function K(e){var t=-1,n=Array(e.size);return e.forEach(function(e,o){n[++t]=[o,e]}),n}function H(e,t){return function(n){return e(t(n))}}function Z(e,t){for(var n=-1,o=e.length,r=0,i=[];++n<o;){var d=e[n];(d===t||d===pe)&&(e[n]=pe,i[r++]=n)}return i}function Y(e){var t=-1,n=Array(e.size);return e.forEach(function(e){n[++t]=e}),n}function Q(e){var t=-1,n=Array(e.size);return e.forEach(function(e){n[++t]=[e,e]}),n}
577function X(e,t,n){for(var o=n-1,r=e.length;++o<r;)if(e[o]===t)return o;return-1}function ee(e,t,n){for(var o=n+1;o--;)if(e[o]===t)return o;return o}function te(e){return q(e)?re(e):Ln(e)}function ne(e){return q(e)?ie(e):m(e)}function oe(e){for(var t=e.length;t--&&kt.test(e.charAt(t)););return t}function re(e){for(var t=cn.lastIndex=0;cn.test(e);)++t;return t}function ie(e){return e.match(cn)||[]}function de(e){return e.match(un)||[]}var ae=200,se="Expected a function",le="__lodash_hash_undefined__",pe="__lodash_placeholder__",_e=1,ce=2,ue=4,he=1,ge=2,fe=1,ye=2,me=4,we=8,be=16,ke=32,xe=64,je=128,ve=256,Se=512,Ae=1,Te=1/0,ze=9007199254740991,Ie=0/0,Oe=4294967295,Ee=[["ary",je],["bind",fe],["bindKey",ye],["curry",we],["curryRight",be],["flip",Se],["partial",ke],["partialRight",xe],["rearg",ve]],Fe="[object Arguments]",Ne="[object Array]",Re="[object Boolean]",We="[object Date]",Ce="[object Error]",Le="[object Function]",Pe="[object GeneratorFunction]",Be="[object Map]",Ue="[object Number]",Ve="[object Object]",Me="[object Promise]",De="[object RegExp]",qe="[object Set]",$e="[object String]",Je="[object Symbol]",Ke="[object WeakMap]",Ge="[object ArrayBuffer]",He="[object DataView]",Ze="[object Float32Array]",Ye="[object Float64Array]",Qe="[object Int8Array]",Xe="[object Int16Array]",et="[object Int32Array]",tt="[object Uint8Array]",nt="[object Uint8ClampedArray]",ot="[object Uint16Array]",rt="[object Uint32Array]",it=/\b__p \+= '';/g,dt=/\b(__p \+=) '' \+/g,at=/(__e\(.*?\)|\b__t\)) \+\n'';/g,st=/&(?:amp|lt|gt|quot|#39);/g,lt=/[&<>"']/g,pt=RegExp(st.source),_t=RegExp(lt.source),ct=/<%-([\s\S]+?)%>/g,ut=/<%([\s\S]+?)%>/g,ht=/<%=([\s\S]+?)%>/g,gt=/\.|\[(?:[^[\]]*|(["'])(?:(?!\1)[^\\]|\\.)*?\1)\]/,ft=/^\w*$/,yt=/[^.[\]]+|\[(?:(-?\d+(?:\.\d+)?)|(["'])((?:(?!\2)[^\\]|\\.)*?)\2)\]|(?=(?:\.|\[\])(?:\.|\[\]|$))/g,mt=/[\\^$.*+?()[\]{}|]/g,wt=RegExp(mt.source),bt=/^\s+/,kt=/\s/,xt=/\{(?:\n\/\* \[wrapped with .+\] \*\/)?\n?/,jt=/\{\n\/\* \[wrapped with (.+)\] \*/,vt=/,? & /,St=/[^\x00-\x2f\x3a-\x40\x5b-\x60\x7b-\x7f]+/g,At=/[()=,{}\[\]\/\s]/,Tt=/\\(\\)?/g,zt=/\$\{([^\\}]*(?:\\.[^\\}]*)*)\}/g,It=/\w*$/,Ot=/^[-+]0x[0-9a-f]+$/i,Et=/^0b[01]+$/i,Ft=/^\[object .+?Constructor\]$/,Nt=/^0o[0-7]+$/i,Rt=/^(?:0|[1-9]\d*)$/,Wt=/[\xc0-\xd6\xd8-\xf6\xf8-\xff\u0100-\u01
5777f]/g,Ct=/($^)/,Lt=/['\n\r\u2028\u2029\\]/g,Pt="\\ud800-\\udfff",Bt="\\u2700-\\u27bf",Ut="a-z\\xdf-\\xf6\\xf8-\\xff",Vt="A-Z\\xc0-\\xd6\\xd8-\\xde",Mt="\\xac\\xb1\\xd7\\xf7"+"\\x00-\\x2f\\x3a-\\x40\\x5b-\\x60\\x7b-\\xbf"+"\\u2000-\\u206f"+" \\t\\x0b\\f\\xa0\\ufeff\\n\\r\\u2028\\u2029\\u1680\\u180e\\u2000\\u2001\\u2002\\u2003\\u2004\\u2005\\u2006\\u2007\\u2008\\u2009\\u200a\\u202f\\u205f\\u3000",Dt="['\u2019]",qt="["+Mt+"]",$t="["+("\\u0300-\\u036f"+"\\ufe20-\\ufe2f"+"\\u20d0-\\u20ff")+"]",Jt="\\d+",Kt="["+Ut+"]",Gt="[^"+Pt+Mt+Jt+Bt+Ut+Vt+"]",Ht="\\ud83c[\\udffb-\\udfff]",Zt="[^"+Pt+"]",Yt="(?:\\ud83c[\\udde6-\\uddff]){2}",Qt="[\\ud800-\\udbff][\\udc00-\\udfff]",Xt="["+Vt+"]",en="(?:"+Kt+"|"+Gt+")",tn="(?:"+Dt+"(?:d|ll|m|re|s|t|ve))?",nn="(?:"+Dt+"(?:D|LL|M|RE|S|T|VE))?",on="(?:"+$t+"|"+Ht+")"+"?",rn="["+"\\ufe0e\\ufe0f"+"]?",dn="(?:"+"\\u200d"+"(?:"+[Zt,Yt,Qt].join("|")+")"+rn+on+")*",an=rn+on+dn,sn="(?:"+["["+Bt+"]",Yt,Qt].join("|")+")"+an,ln="(?:"+[Zt+$t+"?",$t,Yt,Qt,"["+Pt+"]"].join("|")+")",pn=/['â]/g,_n=/[\u0300-\u036f\ufe20-\ufe2f\u20d0-\u20ff]/g,cn=RegExp(Ht+"(?="+Ht+")|"+ln+an,"g"),un=RegExp([Xt+"?"+Kt+"+"+tn+"(?="+[qt,Xt,"$"].join("|")+")","(?:"+Xt+"|"+Gt+")"+"+"+nn+"(?="+[qt,Xt+en,"$"].join("|")+")",Xt+"?"+en+"+"+tn,Xt+"+"+nn,"\\d*(?:1ST|2ND|3RD|(?![123])\\dTH)(?=\\b|[a-z_])","\\d*(?:1st|2nd|3rd|(?![123])\\dth)(?=\\b|[A-Z_])",Jt,sn].join("|"),"g"),hn=/[\u200d\ud800-\udfff\u0300-\u036f\ufe20-\ufe2f\u20d0-\u20ff\ufe0e\ufe0f]/,gn=/[a-z][A-Z]|[A-Z]{2}[a-z]|[0-9][a-zA-Z]|[a-zA-Z][0-9]|[^a-zA-Z0-9 ]/,fn=["Array","Buffer","DataView","Date","Error","Float32Array","Float64Array","Function","Int8Array","Int16Array","Int32Array","Map","Math","Object","Promise","RegExp","Set","String","Symbol","TypeError","Uint8Array","Uint8ClampedArray","Uint16Array","Uint32Array","WeakMap","_","clearTimeout","isFinite","parseInt","setTimeout"],yn=-1,mn={};mn[Ze]=mn[Ye]=mn[Qe]=mn[Xe]=mn[et]=mn[tt]=mn[nt]=mn[ot]=mn[rt]=!0,mn[Fe]=mn[Ne]=mn[Ge]=mn[Re]=mn[He]=mn[We]=mn[Ce]=mn[Le]=mn[Be]=mn[Ue]=mn[Ve]=mn[De]=mn[qe]=mn[$e]=mn[Ke]=!1;var wn={};wn[Fe]=wn[Ne]=wn[Ge]=wn[He]=wn[Re]=wn[We]=wn[Ze]=wn[Ye]=wn[Qe]=wn[Xe]=wn[et]=wn[Be]=wn[Ue]=wn[Ve]=wn[De]=wn[qe]=wn[$e]=wn[Je]=wn[tt]=wn[nt]=wn[ot]=wn[rt]=!0,wn[Ce]=wn[Le]=wn[Ke]=!1;var bn={"\\":"\\","'":"'","\n":"n","\r":"r","\u2028":"u2028","\u2029":"u2029"},kn=parseFloat,xn=parseInt,jn="object"==typeof t&&t&&t.Object===Object&&t,vn="object"==typeof self&&self&&self.Object===Object&&self,Sn=jn||vn||Function("return this")(),An="object"==typeof o&&o&&!o.nodeType&&o,Tn=An&&"object"==typeof n&&n&&!n.nodeType&&n,zn=Tn&&Tn.exports===An,In=zn&&jn.process,On=function(){try{var e=Tn&&Tn.require&&Tn.require("util").types;return e?e:In&&In.binding&&In.binding("util")}catch(t){}}(),En=On&&On.isArrayBuffer,Fn=On&&On.isDate,Nn=On&&On.isMap,Rn=On&&On.isRegExp,Wn=On&&On.isSet,Cn=On&&On.isTypedArray,Ln=A("length"),Pn=z({Ã:"A",Ã:"A",Ã:"A",Ã:"A",Ã:"A",Ã
:"A",à :"a",á:"a",â:"a",ã:"a",ä:"a",Ã¥:"a",Ã:"C",ç:"c",Ã:"D",ð:"d",Ã:"E",Ã:"E",Ã:"E",Ã:"E",è:"e",é:"e",ê:"e",ë:"e",Ã:"I",Ã:"I",Ã:"I",Ã:"I",ì:"i",Ã:"i",î:"i",ï:"i",Ã:"N",ñ:"n",Ã:"O",Ã:"O",Ã:"O",Ã:"O",Ã:"O",Ã:"O",ò:"o",ó:"o",ô:"o",õ:"o",ö:"o",ø:"o",Ã:"U",Ã:"U",Ã:"U",Ã:"U",ù:"u",ú:"u",û:"u",ü:"u",Ã:"Y",ý:"y",ÿ:"y",Ã:"Ae",æ:"ae",Ã:"Th",þ:"th",Ã:"ss",Ä:"A",Ä:"A",Ä:"A",Ä:"a",Ä:"a",Ä
:"a",Ä:"C",Ä:"C",Ä:"C",Ä:"C",Ä:"c",Ä:"c",Ä:"c",Ä:"c",Ä:"D",Ä:"D",Ä:"d",Ä:"d",Ä:"E",Ä:"E",Ä:"E",Ä:"E",Ä:"E",Ä:"e",Ä:"e",Ä:"e",Ä:"e",Ä:"e",Ä:"G",Ä:"G",Ä :"G",Ä¢:"G",Ä:"g",Ä:"g",Ä¡:"g",Ä£:"g",Ĥ:"H",Ħ:"H",Ä¥:"h",ħ:"h",Ĩ:"I",Ī:"I",Ĭ:"I",Ä®:"I",İ:"I",Ä©:"i",Ä«:"i",Ä:"i",į:"i",ı:"i",Ä´:"J",ĵ:"j",Ķ:"K",Ä·:"k",ĸ:"k",Ĺ:"L",Ä»:"L",Ľ:"L",Ä¿:"L",Å:"L",ĺ:"l",ļ:"l",ľ:"l",Å:"l",Å:"l",Å:"N",Å
:"N",Å:"N",Å:"N",Å:"n",Å:"n",Å:"n",Å:"n",Å:"O",Å:"O",Å:"O",Å:"o",Å:"o",Å:"o",Å:"R",Å:"R",Å:"R",Å:"r",Å:"r",Å:"r",Å:"S",Å:"S",Å:"S",Å :"S",Å:"s",Å:"s",Å:"s",Å¡:"s",Å¢:"T",Ť:"T",Ŧ:"T",Å£:"t",Å¥:"t",ŧ:"t",Ũ:"U",Ū:"U",Ŭ:"U",Å®:"U",Ű:"U",Ų:"U",Å©:"u",Å«:"u",Å:"u",ů:"u",ű:"u",ų:"u",Å´:"W",ŵ:"w",Ŷ:"Y",Å·:"y",Ÿ:"Y",Ź:"Z",Å»:"Z",Ž:"Z",ź:"z",ż:"z",ž:"z",IJ:"IJ",ij:"ij",Å:"Oe",Å:"oe",Å:"'n",Å¿:"s"}),Bn=z({"&":"&","<":"<",">":">",'"':""","'":"'"}),Un=z({"&":"&","<":"<",">":">",""":"\"","'":"'"}),Vn=function t(n){function o(e){if(Bi(e)&&!zs(e)&&!(e instanceof z)){if(e instanceof m)return e;if(Id.call(e,"__wrapped__"))return li(e)}return new m(e)}function _(){}function m(e,t){this.__wrapped__=e,this.__actions__=[],this.__chain__=!!t,this.__index__=0,this.__values__=void 0}function z(e){this.__wrapped__=e,this.__actions__=[],this.__dir__=1,this.__filtered__=!1,this.__iteratees__=[],this.__takeCount__=Oe,this.__views__=[]}function G(e){var t=-1,n=null==e?0:e.length;for(this.clear();++t<n;){var o=e[t];this.set(o[0],o[1])}}function X(e){var t=-1,n=null==e?0:e.length;for(this.clear();++t<n;){var o=e[t];this.set(o[0],o[1])}}function re(e){var t=-1,n=null==e?0:e.length;for(this.clear();++t<n;){var o=e[t];this.set(o[0],o[1])}}function ie(e){var t=-1,n=null==e?0:e.length;for(this.__data__=new re;++t<n;)this.add(e[t])}function kt(e){var t=this.__data__=new X(e);this.size=t.size}function St(e,t){var n=zs(e),o=!n&&Ts(e),r=!n&&!o&&Os(e),i=!n&&!o&&!r&&Ws(e),d=n||o||r||i,a=d?F(e.length,xd):[],s=a.length;
577for(var l in e)(t||Id.call(e,l))&&!(d&&("length"==l||r&&("offset"==l||"parent"==l)||i&&("buffer"==l||"byteLength"==l||"byteOffset"==l)||Vr(l,s)))&&a.push(l);return a}function Pt(e){var t=e.length;return t?e[so(0,t-1)]:void 0}function Bt(e,t){return ii(qo(e),Ht(t,0,e.length))}function Ut(e){return ii(qo(e))}function Vt(e,t,n){(void 0===n||Ei(e[t],n))&&(void 0!==n||t in e)||Kt(e,t,n)}function Mt(e,t,n){var o=e[t];Id.call(e,t)&&Ei(o,n)&&(void 0!==n||t in e)||Kt(e,t,n)}function Dt(e,t){for(var n=e.length;n--;)if(Ei(e[n][0],t))return n;return-1}function qt(e,t,n,o){return Ta(e,function(e,r,i){t(o,e,n(e),i)}),o}function $t(e,t){return e&&$o(t,Xi(t),e)}function Jt(e,t){return e&&$o(t,ed(t),e)}function Kt(e,t,n){"__proto__"==t&&Kd?Kd(e,t,{configurable:!0,enumerable:!0,value:n,writable:!0}):e[t]=n}function Gt(e,t){for(var n=-1,o=t.length,r=gd(o);++n<o;)r[n]=null==e?void 0:Yi(e,t[n]);return r}function Ht(e,t,n){return e===e&&(void 0!==n&&(e=e<=n?e:n),void 0!==t&&(e=e>=t?e:t)),e}function Zt(e,t,n,o,r,i){var a=t&_e,s=t&ce,l;if(n&&(l=r?n(e,o,r,i):n(e)),void 0!==l)return l;if(!Pi(e))return e;var p=zs(e);if(!p){var _=Pa(e),c=_==Le||_==Pe;if(Os(e))return Ro(e,a);if(_!=Ve&&_!=Fe&&(!c||r)){if(!wn[_])return r?e:{};l=Pr(e,_,a)}else if(l=s||c?{}:Lr(e),!a)return s?Ko(e,Jt(l,e)):Jo(e,$t(l,e))}else if(l=Cr(e),!a)return qo(e,l);i||(i=new kt);var u=i.get(e);if(u)return u;i.set(e,l),Rs(e)?e.forEach(function(o){l.add(Zt(o,t,n,o,e,i))}):Fs(e)&&e.forEach(function(o,r){l.set(r,Zt(o,t,n,r,e,i))});var h=t&ue?s?Sr:vr:s?ed:Xi,g=p?void 0:h(e);return d(g||e,function(o,r){g&&(r=o,o=e[r]),Mt(l,r,Zt(o,t,n,r,e,i))}),l}function Yt(e){var t=Xi(e);return function(n){return Qt(n,e,t)}}function Qt(e,t,n){var o=n.length;if(null==e)return!o;for(e=bd(e);o--;){var r=n[o],i=t[r],d=e[r];if(void 0===d&&!(r in e)||!i(d))return!1}return!0}function Xt(e,t,n){if("function"!=typeof e)throw new jd(se);return Va(function(){e.apply(void 0,n)},t)}function en(e,t,n,o){var r=-1,i=p,d=!0,a=e.length,s=[],l=t.length;if(!a)return s;n&&(t=u(t,W(n))),o?(i=c,d=!1):t.length>=ae&&(i=L,d=!1,t=new ie(t));outer:for(;++r<a;){var _=e[r],h=null==n?_:n(_);if(_=o||0!==_?_:0,d&&h===h){for(var g=l;g--;)if(t[g]===h)continue outer;s.push(_)}else i(t,h,o)||s.push(_)}return s}function tn(e,t){var n=!0;return Ta(e,function(e,o,r){return n=!!t(e,o,r),n}),n}function nn(e,t,n){for(var o=-1,r=e.length;++o<r;){var i=e[o],d=t(i);if(null!=d&&(void 0===a?d===d&&!Di(d):n(d,a)))var a=d,s=i}return s}function on(e,t,n,o){var r=e.length;for(n=Ji(n),0>n&&(n=-n>r?0:r+n),o=void 0===o||o>r?r:Ji(o),0>o&&(o+=r),o=n>o?0:Ki(o);n<o;)e[n++]=t;return e}function rn(e,t){var n=[];return Ta(e,function(e,o,r){t(e,o,r)&&n.push(e)}),n}function dn(e,t,n,o,r){var i=-1,d=e.length;for(n||(n=Ur),r||(r=[]);++i<d;){var a=e[i];0<t&&n(a)?1<t?dn(a,t-1,n,o,r):h(r,a):!o&&(r[r.length]=a)}return r}function an(e,t){return e&&Ia(e,t,Xi)}function sn(e,t){return e&&Oa(e,t,Xi)}function ln(e,t){return l(t,function(t){return Wi(e[t])})}function cn(e,t){t=Fo(t,e);for(var n=0,o=t.length;null!=e&&n<o;)e=e[di(t[n++])];return n&&n==o?e:void 0}function un(e,t,n){var o=t(e);return zs(e)?o:h(o,n(e))}function hn(e){return null==e?void 0===e?"[object Undefined]":"[object Null]":Jd&&Jd in bd(e)?Fr(e):Qr(e)}function gn(e,t){return e>t}function bn(e,t){return null!=e&&Id.call(e,t)}function jn(e,t){return null!=e&&t in bd(e)}
577function vn(e,t,n){return e>=ia(t,n)&&e<ra(t,n)}function An(e,t,n){for(var o=n?c:p,r=e[0].length,i=e.length,d=i,a=gd(i),s=Infinity,l=[],_;d--;)_=e[d],d&&t&&(_=u(_,W(t))),s=ia(_.length,s),a[d]=!n&&(t||120<=r&&120<=_.length)?new ie(d&&_):void 0;_=e[0];var h=-1,g=a[0];outer:for(;++h<r&&l.length<s;){var f=_[h],y=t?t(f):f;if(f=n||0!==f?f:0,g?!L(g,y):!o(l,y,n)){for(d=i;--d;){var m=a[d];if(m?!L(m,y):!o(e[d],y,n))continue outer}g&&g.push(y),l.push(f)}}return l}function Tn(e,t,n,o){return an(e,function(e,r,i){t(o,n(e),r,i)}),o}function In(e,t,n){t=Fo(t,e),e=ei(e,t);var o=null==e?e:e[di(hi(t))];return null==o?void 0:r(o,e,n)}function On(e){return Bi(e)&&hn(e)==Fe}function Ln(e,t,n,o,r){return e===t||(null!=e&&null!=t&&(Bi(e)||Bi(t))?Mn(e,t,n,o,Ln,r):e!==e&&t!==t)}function Mn(e,t,n,o,r,i){var d=zs(e),a=zs(t),s=d?Ne:Pa(e),l=a?Ne:Pa(t);s=s==Fe?Ve:s,l=l==Fe?Ve:l;var p=s==Ve,_=l==Ve,c=s==l;if(c&&Os(e)){if(!Os(t))return!1;d=!0,p=!1}if(c&&!p)return i||(i=new kt),d||Ws(e)?br(e,t,n,o,r,i):kr(e,t,s,n,o,r,i);if(!(n&he)){var u=p&&Id.call(e,"__wrapped__"),h=_&&Id.call(t,"__wrapped__");
577if(u||h){var g=u?e.value():e,f=h?t.value():t;return i||(i=new kt),r(g,f,n,o,i)}}return!!c&&(i||(i=new kt),xr(e,t,n,o,r,i))}function Dn(e,t,n,o){var r=n.length,i=r,d=!o;if(null==e)return!i;for(e=bd(e);r--;){var a=n[r];if(d&&a[2]?a[1]!==e[a[0]]:!(a[0]in e))return!1}for(;++r<i;){a=n[r];var s=a[0],l=e[s],p=a[1];if(!(d&&a[2])){var _=new kt;if(o)var c=o(l,p,s,e,t,_);if(void 0===c?!Ln(p,l,he|ge,o,_):!c)return!1}else if(void 0===l&&!(s in e))return!1}return!0}function qn(e){if(!Pi(e)||Jr(e))return!1;var t=Wi(e)?Wd:Ft;return t.test(ai(e))}function $n(e){return"function"==typeof e?e:null==e?sd:"object"==typeof e?zs(e)?Yn(e[0],e[1]):Zn(e):cd(e)}function Jn(e){if(!Kr(e))return oa(e);var t=[];for(var n in bd(e))Id.call(e,n)&&"constructor"!=n&&t.push(n);return t}function Kn(e){if(!Pi(e))return Yr(e);var t=Kr(e),n=[];for(var o in e)("constructor"!=o||!t&&Id.call(e,o))&&n.push(o);return n}function Gn(e,t){return e<t}function Hn(e,t){var n=-1,o=Fi(e)?gd(e.length):[];return Ta(e,function(e,r,i){o[++n]=t(e,r,i)}),o}function Zn(e){var t=Or(e);return 1==t.length&&t[0][2]?Hr(t[0][0],t[0][1]):function(n){return n===e||Dn(n,e,t)}}function Yn(e,t){return Dr(e)&&Gr(t)?Hr(di(e),t):function(n){var o=Yi(n,e);return void 0===o&&o===t?Qi(n,e):Ln(t,o,he|ge)}}function Qn(e,t,n,o,r){e===t||Ia(t,function(i,d){if(r||(r=new kt),Pi(i))Xn(e,t,d,n,Qn,o,r);else{var a=o?o(ni(e,d),i,d+"",e,t,r):void 0;void 0===a&&(a=i),Vt(e,d,a)}},ed)}function Xn(e,t,n,o,r,i,d){var a=ni(e,n),s=ni(t,n),l=d.get(s);if(l)return void Vt(e,n,l);var p=i?i(a,s,n+"",e,t,d):void 0,_=void 0===p;if(_){var c=zs(s),u=!c&&Os(s),h=!c&&!u&&Ws(s);p=s,c||u||h?zs(a)?p=a:Ni(a)?p=qo(a):u?(_=!1,p=Ro(s,!0)):h?(_=!1,p=Bo(s,!0)):p=[]:Vi(s)||Ts(s)?(p=a,Ts(a)?p=Hi(a):(!Pi(a)||Wi(a))&&(p=Lr(s))):_=!1}_&&(d.set(s,p),r(p,s,o,i,d),d["delete"](s)),Vt(e,n,p)}function eo(e,t){var n=e.length;if(n)return t+=0>t?n:0,Vr(t,n)?e[t]:void 0}function to(e,t,n){t=t.length?u(t,function(e){return zs(e)?function(t){return cn(t,1===e.length?e[0]:e)}:e}):[sd];var o=-1;t=u(t,W(zr()));var r=Hn(e,function(e){var n=u(t,function(t){return t(e)});return{criteria:n,index:++o,value:e}});return O(r,function(e,t){return Vo(e,t,n)})}function no(e,t){return oo(e,t,function(t,n){return Qi(e,n)})}function oo(e,t,n){for(var o=-1,r=t.length,i={};++o<r;){var d=t[o],a=cn(e,d);n(a,d)&&ho(i,Fo(d,e),a)}return i}function ro(e){return function(t){return cn(t,e)}}function io(e,t,n,o){var r=o?j:x,i=-1,d=t.length,a=e;for(e===t&&(t=qo(t)),n&&(a=u(e,W(n)));++i<d;)for(var s=0,l=t[i],p=n?n(l):l;-1<(s=r(a,p,s,o));)a!==e&&Dd.call(a,s,1),Dd.call(e,s,1);return e}function ao(e,t){for(var n=e?t.length:0,o=n-1,r;n--;)if(r=t[n],n==o||r!==i){var i=r;Vr(r)?Dd.call(e,r,1):vo(e,r)}return e}function so(e,t){return e+Qd(sa()*(t-e+1))}function lo(e,t,n,o){for(var r=-1,i=ra(Yd((t-e)/(n||1)),0),d=gd(i);i--;)d[o?i:++r]=e,e+=n;return d}function po(e,t){var n="";if(!e||1>t||t>ze)return n;do t%2&&(n+=e),t=Qd(t/2),t&&(e+=e);while(t);return n}function _o(e,t){return Ma(Xr(e,t,sd),e+"")}function co(e){return Pt(od(e))}function uo(e,t){var n=od(e);return ii(n,Ht(t,0,n.length))}function ho(e,t,n,o){if(!Pi(e))return e;t=Fo(t,e);for(var r=-1,i=t.length,d=e;null!=d&&++r<i;){var a=di(t[r]),s=n;if("__proto__"===a||"constructor"===a||"prototype"===a)return e;if(r!=i-1){var l=d[a];s=o?o(l,a,d):void 0,void 0===s&&(s=Pi(l)?l:Vr(t[r+1])?[]:{})}Mt(d,a,s),d=d[a]}return e}function go(e){return ii(od(e))}function fo(e,t,n){var o=-1,r=e.length;0>t&&(t=-t>r?0:r+t),n=n>r?r:n,0>n&&(n+=r),r=t>n?0:n-t>>>0,t>>>=0;for(var i=gd(r);++o<r;)i[o]=e[o+t];return i}function yo(e,t){var n;return Ta(e,function(e,o,r){return n=t(e,o,r),!n}),!!n}function mo(e,t,n){var o=0,r=null==e?o:e.length;if("number"==typeof t&&t===t&&r<=Oe>>>1){for(;o<r;){var i=o+r>>>1,d=e[i];null!==d&&!Di(d)&&(n?d<=t:d<t)?o=i+1:r=i}return r}return wo(e,t,sd,n)}function wo(e,t,n,o){var r=0,i=null==e?0:e.length;if(0===i)return 0;t=n(t);for(var d=t!==t,a=null===t,s=Di(t),l=void 0===t;r<i;){var p=Qd((r+i)/2),_=n(e[p]),c=void 0!==_,u=null===_,h=_===_,g=Di(_);if(d)var f=o||h;else f=l?h&&(o||c):a?h&&c&&(o||!u):s?h&&c&&!u&&(o||!g):u||g?!1:o?_<=t:_<t;f?r=p+1:i=p}return ia(i,Oe-1)}function bo(e,t){for(var n=-1,o=e.length,r=0,i=[];++n<o;){var d=e[n],a=t?t(d):d;if(!n||!Ei(a,s)){var s=a;i[r++]=0===d?0:d}}return i}function ko(e){return"number"==typeof e?e:Di(e)?Ie:+e}function xo(e){if("string"==typeof e)return e;if(zs(e))return u(e,xo)+"";if(Di(e))return Sa?Sa.call(e):"";var t=e+"";return"0"==t&&1/e==-Te?"-0":t}function jo(e,t,n){var o=-1,r=p,i=e.length,d=!0,a=[],s=a;if(n)d=!1,r=c;else if(i>=ae){var l=t?null:Ra(e);if(l)return Y(l);d=!1,r=L,s=new ie}else s=t?[]:a;outer:for(;++o<i;){var _=e[o],u=t?t(_):_;if(_=n||0!==_?_:0,d&&u===u){for(var h=s.length;h--;)if(s[h]===u)continue outer;t&&s.push(u),a.push(_)}else r(s,u,n)||(s!==a&&s.push(u),a.push(_))}return a}function vo(e,t){return t=Fo(t,e),e=ei(e,t),null==e||delete e[di(hi(t))]}function So(e,t,n,o){return ho(e,t,n(cn(e,t)),o)}function Ao(e,t,n,o){for(var r=e.length,i=o?r:-1;(o?i--:++i<r)&&t(e[i],i,e););return n?fo(e,o?0:i,o?i+1:r):fo(e,o?i+1:0,o?r:i)}function To(e,t){var n=e;return n instanceof z&&(n=n.value()),g(t,function(e,t){return t.func.apply(t.thisArg,h([e],t.args))},n)}function zo(e,t,n){var o=e.length;if(2>o)return o?jo(e[0]):[];for(var r=-1,i=gd(o);++r<o;)for(var d=e[r],a=-1;++a<o;)a!=r&&(i[r]=en(i[r]||d,e[a],t,n));return jo(dn(i,1),t,n)}
577function Io(e,t,n){for(var o=-1,r=e.length,i=t.length,d={},a;++o<r;)a=o<i?t[o]:void 0,n(d,e[o],a);return d}function Oo(e){return Ni(e)?e:[]}function Eo(e){return"function"==typeof e?e:sd}function Fo(e,t){return zs(e)?e:Dr(e,t)?[e]:Da(Zi(e))}function No(e,t,n){var o=e.length;return n=void 0===n?o:n,!t&&n>=o?e:fo(e,t,n)}function Ro(e,t){if(t)return e.slice();var n=e.length,o=Bd?Bd(n):new e.constructor(n);return e.copy(o),o}function Wo(e){var t=new e.constructor(e.byteLength);return new Pd(t).set(new Pd(e)),t}function Co(e,t){var n=t?Wo(e.buffer):e.buffer;return new e.constructor(n,e.byteOffset,e.byteLength)}function Lo(e){var t=new e.constructor(e.source,It.exec(e));return t.lastIndex=e.lastIndex,t}function Po(e){return va?bd(va.call(e)):{}}function Bo(e,t){var n=t?Wo(e.buffer):e.buffer;return new e.constructor(n,e.byteOffset,e.length)}function Uo(e,t){if(e!==t){var n=void 0!==e,o=null===e,r=e===e,i=Di(e),d=void 0!==t,a=null===t,s=t===t,l=Di(t);if(!a&&!l&&!i&&e>t||i&&d&&s&&!a&&!l||o&&d&&s||!n&&s||!r)return 1;if(!o&&!i&&!l&&e<t||l&&n&&r&&!o&&!i||a&&n&&r||!d&&r||!s)return-1}return 0}function Vo(e,t,n){for(var o=-1,r=e.criteria,i=t.criteria,d=r.length,a=n.length,s;++o<d;)if(s=Uo(r[o],i[o]),s){if(o>=a)return s;var l=n[o];return s*("desc"==l?-1:1)}return e.index-t.index}function Mo(e,t,n,o){for(var r=-1,i=e.length,d=n.length,a=-1,s=t.length,l=ra(i-d,0),p=gd(s+l),_=!o;++a<s;)p[a]=t[a];for(;++r<d;)(_||r<i)&&(p[n[r]]=e[r]);for(;l--;)p[a++]=e[r++];return p}function Do(e,t,n,o){for(var r=-1,i=e.length,d=-1,a=n.length,s=-1,l=t.length,p=ra(i-a,0),_=gd(p+l),c=!o;++r<p;)_[r]=e[r];for(var u=r;++s<l;)_[u+s]=t[s];for(;++d<a;)(c||r<i)&&(_[u+n[d]]=e[r++]);return _}function qo(e,t){var n=-1,o=e.length;for(t||(t=gd(o));++n<o;)t[n]=e[n];return t}function $o(e,t,n,o){var r=!n;n||(n={});for(var i=-1,d=t.length;++i<d;){var a=t[i],s=o?o(n[a],e[a],a,n,e):void 0;void 0===s&&(s=e[a]),r?Kt(n,a,s):Mt(n,a,s)}return n}function Jo(e,t){return $o(e,Ca(e),t)}function Ko(e,t){return $o(e,La(e),t)}function Go(e,t){return function(n,o){var r=zs(n)?i:qt,d=t?t():{};return r(n,e,zr(o,2),d)}}function Ho(e){return _o(function(t,n){var o=-1,r=n.length,i=1<r?n[r-1]:void 0,d=2<r?n[2]:void 0;for(i=3<e.length&&"function"==typeof i?(r--,i):void 0,d&&Mr(n[0],n[1],d)&&(i=3>r?void 0:i,r=1),t=bd(t);++o<r;){var a=n[o];a&&e(t,a,o,i)}return t})}function Zo(e,t){return function(n,o){if(null==n)return n;if(!Fi(n))return e(n,o);for(var r=n.length,i=t?r:-1,d=bd(n);(t?i--:++i<r)&&!1!==o(d[i],i,d););return n}}function Yo(e){return function(t,n,o){for(var r=-1,i=bd(t),d=o(t),a=d.length,s;a--&&(s=d[e?a:++r],!1!==n(i[s],s,i)););return t}}function Qo(e,t,n){function o(){var t=this&&this!==Sn&&this instanceof o?i:e;return t.apply(r?n:this,arguments)}var r=t&fe,i=tr(e);return o}function Xo(e){return function(t){t=Zi(t);var n=q(t)?ne(t):void 0,o=n?n[0]:t.charAt(0),r=n?No(n,1).join(""):t.slice(1);return o[e]()+r}}function er(e){return function(t){return g(dd(id(t).replace(pn,"")),e,"")}}function tr(e){return function(){var t=arguments;switch(t.length){case 0:return new e;case 1:return new e(t[0]);case 2:return new e(t[0],t[1]);case 3:return new e(t[0],t[1],t[2]);case 4:return new e(t[0],t[1],t[2],t[3]);case 5:return new e(t[0],t[1],t[2],t[3],t[4]);case 6:return new e(t[0],t[1],t[2],t[3],t[4],t[5]);case 7:return new e(t[0],t[1],t[2],t[3],t[4],t[5],t[6]);}var n=Aa(e.prototype),o=e.apply(n,t);return Pi(o)?o:n}}function nr(e,t,n){function o(){for(var d=arguments.length,a=gd(d),s=d,l=Tr(o);s--;)a[s]=arguments[s];var p=3>d&&a[0]!==l&&a[d-1]!==l?[]:Z(a,l);if(d-=p.length,d<n)return ur(e,t,ir,o.placeholder,void 0,a,p,void 0,void 0,n-d);var _=this&&this!==Sn&&this instanceof o?i:e;return r(_,this,a)}var i=tr(e);return o}function or(e){return function(t,n,o){var r=bd(t);if(!Fi(t)){var i=zr(n,3);t=Xi(t),n=function(e){return i(r[e],e,r)}}var d=e(t,n,o);return-1<d?r[i?t[d]:d]:void 0}}function rr(e){return jr(function(t){var n=t.length,o=n,r=m.prototype.thru;for(e&&t.reverse();o--;){var i=t[o];if("function"!=typeof i)throw new jd(se);if(r&&!d&&"wrapper"==Ar(i))var d=new m([],!0)}for(o=d?o:n;++o<n;){i=t[o];var a=Ar(i),s="wrapper"==a?Wa(i):void 0;d=s&&$r(s[0])&&s[1]==(je|we|ke|ve)&&!s[4].length&&1==s[9]?d[Ar(s[0])].apply(d,s[3]):1==i.length&&$r(i)?d[a]():d.thru(i)}return function(){var e=arguments,o=e[0];
577if(d&&1==e.length&&zs(o))return d.plant(o).value();for(var r=0,i=n?t[r].apply(this,e):o;++r<n;)i=t[r].call(this,i);return i}})}function ir(e,t,n,o,r,i,d,a,s,l){function p(){for(var y=arguments.length,m=gd(y),w=y;w--;)m[w]=arguments[w];if(h)var b=Tr(p),k=U(m,b);if(o&&(m=Mo(m,o,r,h)),i&&(m=Do(m,i,d,h)),y-=k,h&&y<l){var x=Z(m,b);return ur(e,t,ir,p.placeholder,n,m,x,a,s,l-y)}var j=c?n:this,v=u?j[e]:e;return y=m.length,a?m=ti(m,a):g&&1<y&&m.reverse(),_&&s<y&&(m.length=s),this&&this!==Sn&&this instanceof p&&(v=f||tr(v)),v.apply(j,m)}var _=t&je,c=t&fe,u=t&ye,h=t&(we|be),g=t&Se,f=u?void 0:tr(e);return p}function dr(e,t){return function(n,o){return Tn(n,e,t(o),{})}}function ar(e,t){return function(n,o){var r;if(void 0===n&&void 0===o)return t;if(void 0!==n&&(r=n),void 0!==o){if(void 0===r)return o;"string"==typeof n||"string"==typeof o?(n=xo(n),o=xo(o)):(n=ko(n),o=ko(o)),r=e(n,o)}return r}}function sr(e){return jr(function(t){return t=u(t,W(zr())),_o(function(n){var o=this;return e(t,function(e){return r(e,o,n)})})})}function lr(e,t){t=void 0===t?" ":xo(t);var n=t.length;if(2>n)return n?po(t,e):t;var o=po(t,Yd(e/te(t)));return q(t)?No(ne(o),0,e).join(""):o.slice(0,e)}function pr(e,t,n,o){function i(){for(var t=-1,s=arguments.length,l=-1,p=o.length,_=gd(p+s),c=this&&this!==Sn&&this instanceof i?a:e;++l<p;)_[l]=o[l];for(;s--;)_[l++]=arguments[++t];return r(c,d?n:this,_)}var d=t&fe,a=tr(e);return i}function _r(e){return function(t,n,o){return o&&"number"!=typeof o&&Mr(t,n,o)&&(n=o=void 0),t=$i(t),void 0===n?(n=t,t=0):n=$i(n),o=void 0===o?t<n?1:-1:$i(o),lo(t,n,o,e)}}function cr(e){return function(t,n){return"string"==typeof t&&"string"==typeof n||(t=Gi(t),n=Gi(n)),e(t,n)}}function ur(e,t,n,o,r,i,d,a,s,l){var p=t&we,_=p?d:void 0,c=p?void 0:d,u=p?i:void 0,h=p?void 0:i;t|=p?ke:xe,t&=~(p?xe:ke),t&me||(t&=~(fe|ye));var g=[e,t,r,u,_,h,c,a,s,l],f=n.apply(void 0,g);return $r(e)&&Ua(f,g),f.placeholder=o,oi(f,e,t)}function hr(e){var t=wd[e];return function(e,n){if(e=Gi(e),n=null==n?0:ia(Ji(n),292),n&&ta(e)){var o=(Zi(e)+"e").split("e"),r=t(o[0]+"e"+(+o[1]+n));return o=(Zi(r)+"e").split("e"),+(o[0]+"e"+(+o[1]-n))}return t(e)}}function gr(e){return function(t){var n=Pa(t);return n==Be?K(t):n==qe?Q(t):N(t,e(t))}}function fr(e,t,n,o,r,i,d,a){var s=t&ye;if(!s&&"function"!=typeof e)throw new jd(se);var l=o?o.length:0;if(l||(t&=~(ke|xe),o=r=void 0),d=void 0===d?d:ra(Ji(d),0),a=void 0===a?a:Ji(a),l-=r?r.length:0,t&xe){var p=o,_=r;o=r=void 0}var c=s?void 0:Wa(e),u=[e,t,n,o,r,p,_,i,d,a];if(c&&Zr(u,c),e=u[0],t=u[1],n=u[2],o=u[3],r=u[4],a=u[9]=void 0===u[9]?s?0:e.length:ra(u[9]-l,0),!a&&t&(we|be)&&(t&=~(we|be)),!t||t==fe)var h=Qo(e,t,n);else h=t==we||t==be?nr(e,t,a):t!=ke&&t!=(fe|ke)||r.length?ir.apply(void 0,u):pr(e,t,n,o);var g=c?Ea:Ua;return oi(g(h,u),e,t)}function yr(e,t,n,o){return void 0===e||Ei(e,Ad[n])&&!Id.call(o,n)?t:e}function mr(e,t,n,o,r,i){return Pi(e)&&Pi(t)&&(i.set(t,e),Qn(e,t,void 0,mr,i),i["delete"](t)),e}function wr(e){return Vi(e)?void 0:e}function br(e,t,n,o,r,i){var d=n&he,a=e.length,s=t.length;if(a!=s&&!(d&&s>a))return!1;var l=i.get(e),p=i.get(t);if(l&&p)return l==t&&p==e;var _=-1,c=!0,u=n&ge?new ie:void 0;for(i.set(e,t),i.set(t,e);++_<a;){var h=e[_],g=t[_];if(o)var f=d?o(g,h,_,t,e,i):o(h,g,_,e,t,i);if(void 0!==f){if(f)continue;c=!1;break}if(u){if(!y(t,function(e,t){if(!L(u,t)&&(h===e||r(h,e,n,o,i)))return u.push(t)})){c=!1;break}}else if(!(h===g||r(h,g,n,o,i))){c=!1;break}}return i["delete"](e),i["delete"](t),c}function kr(e,t,n,o,r,i,d){switch(n){case He:if(e.byteLength!=t.byteLength||e.byteOffset!=t.byteOffset)return!1;e=e.buffer,t=t.buffer;case Ge:return e.byteLength==t.byteLength&&i(new Pd(e),new Pd(t));case Re:case We:case Ue:return Ei(+e,+t);case Ce:return e.name==t.name&&e.message==t.message;case De:case $e:return e==t+"";case Be:var a=K;case qe:var s=o&he;if(a||(a=Y),e.size!=t.size&&!s)return!1;var l=d.get(e);if(l)return l==t;o|=ge,d.set(e,t);var p=br(a(e),a(t),o,r,i,d);return d["delete"](e),p;case Je:if(va)return va.call(e)==va.call(t);}return!1}function xr(e,t,n,o,r,i){var d=n&he,a=vr(e),s=a.length,l=vr(t),p=l.length;if(s!=p&&!d)return!1;for(var _=s,c;_--;)if(c=a[_],d?!(c in t):!Id.call(t,c))return!1;var u=i.get(e),h=i.get(t);if(u&&h)return u==t&&h==e;var g=!0;i.set(e,t),i.set(t,e);for(var f=d;++_<s;){c=a[_];var y=e[c],m=t[c];if(o)var w=d?o(m,y,c,t,e,i):o(y,m,c,e,t,i);if(void 0===w?!(y===m||r(y,m,n,o,i)):!w){g=!1;break}f||(f="constructor"==c)}if(g&&!f){var b=e.constructor,k=t.constructor;b!=k&&"constructor"in e&&"constructor"in t&&!("function"==typeof b&&b instanceof b&&"function"==typeof k&&k instanceof k)&&(g=!1)}return i["delete"](e),i["delete"](t),g}function jr(e){return Ma(Xr(e,void 0,ci),e+"")}function vr(e){return un(e,Xi,Ca)}function Sr(e){return un(e,ed,La)}function Ar(e){for(var t=e.name+"",n=ya[t],o=Id.call(ya,t)?n.length:0;o--;){var r=n[o],i=r.func;if(null==i||i==e)return r.name}return t}function Tr(e){var t=Id.call(o,"placeholder")?o:e;return t.placeholder}function zr(){var e=o.iteratee||ld;return e=e===ld?$n:e,arguments.length?e(arguments[0],arguments[1]):e}function Ir(e,t){var n=e.__data__;return qr(t)?n["string"==typeof t?"string":"hash"]:n.map}function Or(e){for(var t=Xi(e),n=t.length;n--;){var o=t[n],r=e[o];t[n]=[o,r,Gr(r)]}return t}function Er(e,t){var n=D(e,t);return qn(n)?n:void 0}function Fr(e){var t=Id.call(e,Jd),n=e[Jd];try{e[Jd]=void 0}catch(t){}var o=Fd.call(e);return t?e[Jd]=n:delete e[Jd],o}function Nr(e,t,n){for(var o=-1,r=n.length;++o<r;){var i=n[o],d=i.size;switch(i.type){case"drop":e+=d;break;case"dropRight":t-=d;break;case"take":t=ia(t,e+d);break;case"takeRight":e=ra(e,t-d);}}return{start:e,end:t}}function Rr(e){var t=e.match(jt);return t?t[1].split(vt):[]}function Wr(e,t,n){t=Fo(t,e);for(var o=-1,r=t.length,i=!1,d;++o<r&&(d=di(t[o]),!!(i=null!=e&&n(e,d)));)e=e[d];return i||++o!=r?i:(r=null==e?0:e.length,!!r&&Li(r)&&Vr(d,r)&&(zs(e)||Ts(e)))}function Cr(e){var t=e.length,n=new e.constructor(t);return t&&"string"==typeof e[0]&&Id.call(e,"index")&&(n.index=e.index,n.input=e.input),n}function Lr(e){return"function"!=typeof e.constructor||Kr(e)?{}:Aa(Ud(e))}function Pr(e,t,n){var o=e.constructor;return t===Ge?Wo(e):t===Re||t===We?new o(+e):t===He?Co(e,n):t===Ze||t===Ye||t===Qe||t===Xe||t===et||t===tt||t===nt||t===ot||t===rt?Bo(e,n):t===Be?new o:t===Ue||t===$e?new o(e):t===De?Lo(e):t===qe?new o:t===Je?Po(e):void 0}function Br(e,t){var n=t.length;if(!n)return e;var o=n-1;return t[o]=(1<n?"& ":"")+t[o],t=t.join(2<n?", ":" "),e.replace(xt,"{\n/* [wrapped with "+t+"] */\n")}
577function Ur(e){return zs(e)||Ts(e)||!!(qd&&e&&e[qd])}function Vr(e,t){var n=typeof e;return t=null==t?ze:t,!!t&&("number"==n||"symbol"!=n&&Rt.test(e))&&-1<e&&0==e%1&&e<t}function Mr(e,t,n){if(!Pi(n))return!1;var o=typeof t;return("number"==o?!!(Fi(n)&&Vr(t,n.length)):!!("string"==o&&t in n))&&Ei(n[t],e)}function Dr(e,t){if(zs(e))return!1;var n=typeof e;return"number"==n||"symbol"==n||"boolean"==n||null==e||Di(e)||ft.test(e)||!gt.test(e)||null!=t&&e in bd(t)}function qr(e){var t=typeof e;return"string"==t||"number"==t||"symbol"==t||"boolean"==t?"__proto__"!==e:null===e}function $r(e){var t=Ar(e),n=o[t];if("function"!=typeof n||!(t in z.prototype))return!1;if(e===n)return!0;var r=Wa(n);return!!r&&e===r[0]}function Jr(e){return!!Ed&&Ed in e}function Kr(e){var t=e&&e.constructor,n="function"==typeof t&&t.prototype||Ad;return e===n}function Gr(e){return e===e&&!Pi(e)}function Hr(e,t){return function(n){return null!=n&&n[e]===t&&(void 0!==t||e in bd(n))}}function Zr(e,t){var n=e[1],o=t[1],r=n|o,i=r<(fe|ye|je),d=o==je&&n==we||o==je&&n==ve&&e[7].length<=t[8]||o==(je|ve)&&t[7].length<=t[8]&&n==we;if(!(i||d))return e;o&fe&&(e[2]=t[2],r|=n&fe?0:me);var a=t[3];if(a){var s=e[3];e[3]=s?Mo(s,a,t[4]):a,e[4]=s?Z(e[3],pe):t[4]}return a=t[5],a&&(s=e[5],e[5]=s?Do(s,a,t[6]):a,e[6]=s?Z(e[5],pe):t[6]),a=t[7],a&&(e[7]=a),o&je&&(e[8]=null==e[8]?t[8]:ia(e[8],t[8])),null==e[9]&&(e[9]=t[9]),e[0]=t[0],e[1]=r,e}function Yr(e){var t=[];if(null!=e)for(var n in bd(e))t.push(n);return t}function Qr(e){return Fd.call(e)}function Xr(e,t,n){return t=ra(void 0===t?e.length-1:t,0),function(){for(var o=arguments,i=-1,d=ra(o.length-t,0),a=gd(d);++i<d;)a[i]=o[t+i];i=-1;for(var s=gd(t+1);++i<t;)s[i]=o[i];return s[t]=n(a),r(e,this,s)}}function ei(e,t){return 2>t.length?e:cn(e,fo(t,0,-1))}function ti(e,t){for(var n=e.length,o=ia(t.length,n),r=qo(e),i;o--;)i=t[o],e[o]=Vr(i,n)?r[i]:void 0;return e}function ni(e,t){return"constructor"===t&&"function"==typeof e[t]||"__proto__"==t?void 0:e[t]}function oi(e,t,n){var o=t+"";return Ma(e,Br(o,si(Rr(o),n)))}function ri(e){var t=0,n=0;return function(){var o=da(),r=16-(o-n);if(n=o,!(0<r))t=0;else if(++t>=800)return arguments[0];return e.apply(void 0,arguments)}}function ii(e,t){var n=-1,o=e.length;for(t=void 0===t?o:t;++n<t;){var r=so(n,o-1),i=e[r];e[r]=e[n],e[n]=i}return e.length=t,e}function di(e){if("string"==typeof e||Di(e))return e;var t=e+"";return"0"==t&&1/e==-Te?"-0":t}function ai(e){if(null!=e){try{return zd.call(e)}catch(t){}try{return e+""}catch(t){}}return""}function si(e,t){return d(Ee,function(n){var o="_."+n[0];t&n[1]&&!p(e,o)&&e.push(o)}),e.sort()}function li(e){if(e instanceof z)return e.clone();var t=new m(e.__wrapped__,e.__chain__);return t.__actions__=qo(e.__actions__),t.__index__=e.__index__,t.__values__=e.__values__,t}function pi(e,t,n){var o=null==e?0:e.length;if(!o)return-1;var r=null==n?0:Ji(n);return 0>r&&(r=ra(o+r,0)),k(e,zr(t,3),r)}function _i(e,t,n){var o=null==e?0:e.length;if(!o)return-1;var r=o-1;return void 0!==n&&(r=Ji(n),r=0>n?ra(o+r,0):ia(r,o-1)),k(e,zr(t,3),r,!0)}function ci(e){var t=null==e?0:e.length;return t?dn(e,1):[]}function ui(e){return e&&e.length?e[0]:void 0}function hi(e){var t=null==e?0:e.length;return t?e[t-1]:void 0}function gi(e,t){return e&&e.length&&t&&t.length?io(e,t):e}function fi(e){return null==e?e:la.call(e)}function yi(e){if(!(e&&e.length))return[];var t=0;return e=l(e,function(e){if(Ni(e))return t=ra(e.length,t),!0}),F(t,function(t){return u(e,A(t))})}function mi(e,t){if(!(e&&e.length))return[];var n=yi(e);return null==t?n:u(n,function(e){return r(t,void 0,e)})}function wi(e){var t=o(e);return t.__chain__=!0,t}function bi(e,t){return t(e)}function ki(e,t){var n=zs(e)?d:Ta;return n(e,zr(t,3))}function xi(e,t){var n=zs(e)?a:za;return n(e,zr(t,3))}function ji(e,t){var n=zs(e)?u:Hn;return n(e,zr(t,3))}function vi(e,t,n){return t=n?void 0:t,t=e&&null==t?e.length:t,fr(e,je,void 0,void 0,void 0,void 0,t)}function Si(e,t){var n;if("function"!=typeof t)throw new jd(se);return e=Ji(e),function(){return 0<--e&&(n=t.apply(this,arguments)),1>=e&&(t=void 0),n}}function Ai(e,t,n){t=n?void 0:t;var o=fr(e,we,void 0,void 0,void 0,void 0,void 0,t);return o.placeholder=Ai.placeholder,o}
577function Ti(e,t,n){t=n?void 0:t;var o=fr(e,be,void 0,void 0,void 0,void 0,void 0,t);return o.placeholder=Ti.placeholder,o}function zi(e,t,n){function o(t){var n=h,o=g;return h=g=void 0,p=t,y=e.apply(o,n),y}function r(e){return p=e,m=Va(a,t),_?o(e):y}function i(e){var n=e-w,o=e-p,r=t-n;return c?ia(r,f-o):r}function d(e){var n=e-w,o=e-p;return void 0==w||n>=t||0>n||c&&o>=f}function a(){var e=fs();return d(e)?s(e):void(m=Va(a,i(e)))}function s(e){return(m=void 0,u&&h)?o(e):(h=g=void 0,y)}function l(){var e=fs(),n=d(e);if(h=arguments,g=this,w=e,n){if(void 0===m)return r(w);if(c)return Na(m),m=Va(a,t),o(w)}return void 0===m&&(m=Va(a,t)),y}var p=0,_=!1,c=!1,u=!0,h,g,f,y,m,w;if("function"!=typeof e)throw new jd(se);return t=Gi(t)||0,Pi(n)&&(_=!!n.leading,c="maxWait"in n,f=c?ra(Gi(n.maxWait)||0,t):f,u="trailing"in n?!!n.trailing:u),l.cancel=function(){void 0!==m&&Na(m),p=0,h=w=g=m=void 0},l.flush=function(){return void 0===m?y:s(fs())},l}function Ii(e,t){if("function"!=typeof e||null!=t&&"function"!=typeof t)throw new jd(se);var n=function(){var o=arguments,r=t?t.apply(this,o):o[0],i=n.cache;if(i.has(r))return i.get(r);var d=e.apply(this,o);return n.cache=i.set(r,d)||i,d};return n.cache=new(Ii.Cache||re),n}function Oi(e){if("function"!=typeof e)throw new jd(se);return function(){var t=arguments;switch(t.length){case 0:return!e.call(this);case 1:return!e.call(this,t[0]);case 2:return!e.call(this,t[0],t[1]);case 3:return!e.call(this,t[0],t[1],t[2]);}return!e.apply(this,t)}}function Ei(e,t){return e===t||e!==e&&t!==t}function Fi(e){return null!=e&&Li(e.length)&&!Wi(e)}function Ni(e){return Bi(e)&&Fi(e)}function Ri(e){if(!Bi(e))return!1;var t=hn(e);return t==Ce||t=="[object DOMException]"||"string"==typeof e.message&&"string"==typeof e.name&&!Vi(e)}function Wi(e){if(!Pi(e))return!1;var t=hn(e);return t==Le||t==Pe||t=="[object AsyncFunction]"||t=="[object Proxy]"}function Ci(e){return"number"==typeof e&&e==Ji(e)}function Li(e){return"number"==typeof e&&-1<e&&0==e%1&&e<=ze}function Pi(e){var t=typeof e;return null!=e&&("object"==t||"function"==t)}function Bi(e){return null!=e&&"object"==typeof e}function Ui(e){return"number"==typeof e||Bi(e)&&hn(e)==Ue}function Vi(e){if(!Bi(e)||hn(e)!=Ve)return!1;var t=Ud(e);if(null===t)return!0;var n=Id.call(t,"constructor")&&t.constructor;return"function"==typeof n&&n instanceof n&&zd.call(n)==Nd}function Mi(e){return"string"==typeof e||!zs(e)&&Bi(e)&&hn(e)==$e}function Di(e){return"symbol"==typeof e||Bi(e)&&hn(e)==Je}function qi(e){if(!e)return[];if(Fi(e))return Mi(e)?ne(e):qo(e);if($d&&e[$d])return J(e[$d]());var t=Pa(e),n=t==Be?K:t==qe?Y:od;return n(e)}function $i(e){if(!e)return 0===e?e:0;if(e=Gi(e),e===Te||e===-Te){var t=0>e?-1:1;return t*1.7976931348623157e308}return e===e?e:0}function Ji(e){var t=$i(e),n=t%1;return t===t?n?t-n:t:0}function Ki(e){return e?Ht(Ji(e),0,Oe):0}function Gi(e){if("number"==typeof e)return e;if(Di(e))return Ie;if(Pi(e)){var t="function"==typeof e.valueOf?e.valueOf():e;e=Pi(t)?t+"":t}if("string"!=typeof e)return 0===e?e:+e;e=R(e);var n=Et.test(e);return n||Nt.test(e)?xn(e.slice(2),n?2:8):Ot.test(e)?Ie:+e}function Hi(e){return $o(e,ed(e))}function Zi(e){return null==e?"":xo(e)}function Yi(e,t,n){var o=null==e?void 0:cn(e,t);return void 0===o?n:o}function Qi(e,t){return null!=e&&Wr(e,t,jn)}function Xi(e){return Fi(e)?St(e):Jn(e)}function ed(e){return Fi(e)?St(e,!0):Kn(e)}function td(e,t){if(null==e)return{};var n=u(Sr(e),function(e){return[e]});return t=zr(t),oo(e,n,function(e,n){return t(e,n[0])})}function nd(e,t,n){t=Fo(t,e);var o=-1,r=t.length;for(r||(r=1,e=void 0);++o<r;){var i=null==e?void 0:e[di(t[o])];void 0===i&&(o=r,i=n),e=Wi(i)?i.call(e):i}return e}function od(e){return null==e?[]:C(e,Xi(e))}function rd(e){return al(Zi(e).toLowerCase())}function id(e){return e=Zi(e),e&&e.replace(Wt,Pn).replace(_n,"")}function dd(e,t,n){return e=Zi(e),t=n?void 0:t,void 0===t?$(e)?de(e):w(e):e.match(t)||[]}function ad(e){return function(){return e}}function sd(e){return e}function ld(e){return $n("function"==typeof e?e:Zt(e,_e))}function pd(e,t,n){var o=Xi(t),r=ln(t,o);null!=n||Pi(t)&&(r.length||!o.length)||(n=t,t=e,e=this,r=ln(t,Xi(t)));var i=!(Pi(n)&&"chain"in n)||!!n.chain,a=Wi(e);return d(r,function(n){var o=t[n];e[n]=o,a&&(e.prototype[n]=function(){var t=this.__chain__;if(i||t){var n=e(this.__wrapped__),r=n.__actions__=qo(this.__actions__);return r.push({func:o,args:arguments,thisArg:e}),n.__chain__=t,n}return o.apply(e,h([this.value()],arguments))})}),e}function _d(){}function cd(e){return Dr(e)?A(di(e)):ro(e)}function ud(){return[]}function hd(){return!1}n=null==n?Sn:Vn.defaults(Sn.Object(),n,Vn.pick(Sn,fn));var gd=n.Array,fd=n.Date,yd=n.Error,md=n.Function,wd=n.Math,bd=n.Object,kd=n.RegExp,xd=n.String,jd=n.TypeError,vd=gd.prototype,Sd=md.prototype,Ad=bd.prototype,Td=n["__core-js_shared__"],zd=Sd.toString,Id=Ad.hasOwnProperty,Od=0,Ed=function(){var e=/[^.]+$/.exec(Td&&Td.keys&&Td.keys.IE_PROTO||"");
577return e?"Symbol(src)_1."+e:""}(),Fd=Ad.toString,Nd=zd.call(bd),Rd=Sn._,Wd=kd("^"+zd.call(Id).replace(mt,"\\$&").replace(/hasOwnProperty|(function).*?(?=\\\()| for .+?(?=\\\])/g,"$1.*?")+"$"),Cd=zn?n.Buffer:void 0,Ld=n.Symbol,Pd=n.Uint8Array,Bd=Cd?Cd.allocUnsafe:void 0,Ud=H(bd.getPrototypeOf,bd),Vd=bd.create,Md=Ad.propertyIsEnumerable,Dd=vd.splice,qd=Ld?Ld.isConcatSpreadable:void 0,$d=Ld?Ld.iterator:void 0,Jd=Ld?Ld.toStringTag:void 0,Kd=function(){try{var e=Er(bd,"defineProperty");return e({},"",{}),e}catch(t){}}(),Gd=n.clearTimeout!==Sn.clearTimeout&&n.clearTimeout,Hd=fd&&fd.now!==Sn.Date.now&&fd.now,Zd=n.setTimeout!==Sn.setTimeout&&n.setTimeout,Yd=wd.ceil,Qd=e,Xd=bd.getOwnPropertySymbols,ea=Cd?Cd.isBuffer:void 0,ta=n.isFinite,na=vd.join,oa=H(bd.keys,bd),ra=T,ia=V,da=fd.now,aa=n.parseInt,sa=wd.random,la=vd.reverse,pa=Er(n,"DataView"),_a=Er(n,"Map"),ca=Er(n,"Promise"),ua=Er(n,"Set"),ha=Er(n,"WeakMap"),ga=Er(bd,"create"),fa=ha&&new ha,ya={},ma=ai(pa),wa=ai(_a),ba=ai(ca),ka=ai(ua),xa=ai(ha),ja=Ld?Ld.prototype:void 0,va=ja?ja.valueOf:void 0,Sa=ja?ja.toString:void 0,Aa=function(){function e(){}return function(t){if(!Pi(t))return{};if(Vd)return Vd(t);e.prototype=t;var n=new e;return e.prototype=void 0,n}}();o.templateSettings={escape:ct,evaluate:ut,interpolate:ht,variable:"",imports:{_:o}},o.prototype=_.prototype,o.prototype.constructor=o,m.prototype=Aa(_.prototype),m.prototype.constructor=m,z.prototype=Aa(_.prototype),z.prototype.constructor=z,G.prototype.clear=function(){this.__data__=ga?ga(null):{},this.size=0},G.prototype["delete"]=function(e){var t=this.has(e)&&delete this.__data__[e];return this.size-=t?1:0,t},G.prototype.get=function(e){var t=this.__data__;if(ga){var n=t[e];return n===le?void 0:n}return Id.call(t,e)?t[e]:void 0},G.prototype.has=function(e){var t=this.__data__;return ga?void 0!==t[e]:Id.call(t,e)},G.prototype.set=function(e,t){var n=this.__data__;return this.size+=this.has(e)?0:1,n[e]=ga&&void 0===t?le:t,this},X.prototype.clear=function(){this.__data__=[],this.size=0},X.prototype["delete"]=function(e){var t=this.__data__,n=Dt(t,e);if(0>n)return!1;var o=t.length-1;return n==o?t.pop():Dd.call(t,n,1),--this.size,!0},X.prototype.get=function(e){var t=this.__data__,n=Dt(t,e);return 0>n?void 0:t[n][1]},X.prototype.has=function(e){return-1<Dt(this.__data__,e)},X.prototype.set=function(e,t){var n=this.__data__,o=Dt(n,e);return 0>o?(++this.size,n.push([e,t])):n[o][1]=t,this},re.prototype.clear=function(){this.size=0,this.__data__={hash:new G,map:new(_a||X),string:new G}},re.prototype["delete"]=function(e){var t=Ir(this,e)["delete"](e);return this.size-=t?1:0,t},re.prototype.get=function(e){return Ir(this,e).get(e)},re.prototype.has=function(e){return Ir(this,e).has(e)},re.prototype.set=function(e,t){var n=Ir(this,e),o=n.size;return n.set(e,t),this.size+=n.size==o?0:1,this},ie.prototype.add=ie.prototype.push=function(e){return this.__data__.set(e,le),this},ie.prototype.has=function(e){return this.__data__.has(e)},kt.prototype.clear=function(){this.__data__=new X,this.size=0},kt.prototype["delete"]=function(e){var t=this.__data__,n=t["delete"](e);return this.size=t.size,n},kt.prototype.get=function(e){return this.__data__.get(e)},kt.prototype.has=function(e){return this.__data__.has(e)},kt.prototype.set=function(e,t){var n=this.__data__;if(n instanceof X){var o=n.__data__;if(!_a||o.length<ae-1)return o.push([e,t]),this.size=++n.size,this;n=this.__data__=new re(o)}return n.set(e,t),this.size=n.size,this};var Ta=Zo(an),za=Zo(sn,!0),Ia=Yo(),Oa=Yo(!0),Ea=fa?function(e,t){return fa.set(e,t),e}:sd,Fa=Kd?function(e,t){return Kd(e,"toString",{configurable:!0,enumerable:!1,value:ad(t),writable:!0})}:sd,Na=Gd||function(e){return Sn.clearTimeout(e)},Ra=ua&&1/Y(new ua([,-0]))[1]==Te?function(e){return new ua(e)}:_d,Wa=fa?function(e){return fa.get(e)}:_d,Ca=Xd?function(e){return null==e?[]:(e=bd(e),l(Xd(e),function(t){return Md.call(e,t)}))}:ud,La=Xd?function(e){for(var t=[];e;)h(t,Ca(e)),e=Ud(e);return t}:ud,Pa=hn;(pa&&Pa(new pa(new ArrayBuffer(1)))!=He||_a&&Pa(new _a)!=Be||ca&&Pa(ca.resolve())!=Me||ua&&Pa(new ua)!=qe||ha&&Pa(new ha)!=Ke)&&(Pa=function(e){var t=hn(e),n=t==Ve?e.constructor:void 0,o=n?ai(n):"";if(o)switch(o){case ma:return He;case wa:return Be;case ba:return Me;case ka:return qe;case xa:return Ke;}return t});var Ba=Td?Wi:hd,Ua=ri(Ea),Va=Zd||function(e,t){return Sn.setTimeout(e,t)},Ma=ri(Fa),Da=function(e){var t=Ii(e,function(e){return n.size===500&&n.clear(),e}),n=t.cache;return t}(function(e){var t=[];return 46===e.charCodeAt(0)&&t.push(""),e.replace(yt,function(e,n,o,r){t.push(o?r.replace(Tt,"$1"):n||e)}),t}),qa=_o(function(e,t){return Ni(e)?en(e,dn(t,1,Ni,!0)):[]}),$a=_o(function(e,t){var n=hi(t);return Ni(n)&&(n=void 0),Ni(e)?en(e,dn(t,1,Ni,!0),zr(n,2)):[]}),Ja=_o(function(e,t){var n=hi(t);return Ni(n)&&(n=void 0),Ni(e)?en(e,dn(t,1,Ni,!0),void 0,n):[]}),Ka=_o(function(e){var t=u(e,Oo);return t.length&&t[0]===e[0]?An(t):[]}),Ga=_o(function(e){var t=hi(e),n=u(e,Oo);return t===hi(n)?t=void 0:n.pop(),n.length&&n[0]===e[0]?An(n,zr(t,2)):[]}),Ha=_o(function(e){var t=hi(e),n=u(e,Oo);return t="function"==typeof t?t:void 0,t&&n.pop(),n.length&&n[0]===e[0]?An(n,void 0,t):[]}),Za=_o(gi),Ya=jr(function(e,t){var n=null==e?0:e.length,o=Gt(e,t);return ao(e,u(t,function(e){return Vr(e,n)?+e:e}).sort(Uo)),o}),Qa=_o(function(e){return jo(dn(e,1,Ni,!0))}),Xa=_o(function(e){var t=hi(e);return Ni(t)&&(t=void 0),jo(dn(e,1,Ni,!0),zr(t,2))}),es=_o(function(e){var t=hi(e);return t="function"==typeof t?t:void 0,jo(dn(e,1,Ni,!0),void 0,t)}),ts=_o(function(e,t){return Ni(e)?en(e,t):[]}),ns=_o(function(e){return zo(l(e,Ni))}),os=_o(function(e){var t=hi(e);return Ni(t)&&(t=void 0),zo(l(e,Ni),zr(t,2))}),rs=_o(function(e){var t=hi(e);return t="function"==typeof t?t:void 0,zo(l(e,Ni),void 0,t)}),is=_o(yi),ds=_o(function(e){var t=e.length,n=1<t?e[t-1]:void 0;return n="function"==typeof n?(e.pop(),n):void 0,mi(e,n)}),as=jr(function(e){var t=e.length,n=t?e[0]:0,o=this.__wrapped__,r=function(t){return Gt(t,e)};return 1<t||this.__actions__.length||!(o instanceof z)||!Vr(n)?this.thru(r):(o=o.slice(n,+n+(t?1:0)),o.__actions__.push({func:bi,args:[r],thisArg:void 0}),new m(o,this.__chain__).thru(function(e){return t&&!e.length&&e.push(void 0),e}))}),ss=Go(function(e,t,n){Id.call(e,n)?++e[n]:Kt(e,n,1)}),ls=or(pi),ps=or(_i),_s=Go(function(e,t,n){Id.call(e,n)?e[n].push(t):Kt(e,n,[t])}),cs=_o(function(e,t,n){var o=-1,i=Fi(e)?gd(e.length):[];return Ta(e,function(e){i[++o]="function"==typeof t?r(t,e,n):In(e,t,n)}),i}),us=Go(function(e,t,n){Kt(e,n,t)}),hs=Go(function(e,t,n){e[n?0:1].push(t)},function(){return[[],[]]}),gs=_o(function(e,t){if(null==e)return[];var n=t.length;return 1<n&&Mr(e,t[0],t[1])?t=[]:2<n&&Mr(t[0],t[1],t[2])&&(t=[t[0]]),to(e,dn(t,1),[])}),fs=Hd||function(){return Sn.Date.now()},ys=_o(function(e,t,n){var o=fe;if(n.length){var r=Z(n,Tr(ys));o|=ke}return fr(e,o,t,n,r)}),ms=_o(function(e,t,n){var o=fe|ye;if(n.length){var r=Z(n,Tr(ms));o|=ke}return fr(t,o,e,n,r)}),ws=_o(function(e,t){return Xt(e,1,t)}),bs=_o(function(e,t,n){return Xt(e,Gi(t)||0,n)});Ii.Cache=re;var ks=_o(function(e,t){t=1==t.length&&zs(t[0])?u(t[0],W(zr())):u(dn(t,1),W(zr()));var n=t.length;return _o(function(o){for(var i=-1,d=ia(o.length,n);++i<d;)o[i]=t[i].call(this,o[i]);return r(e,this,o)})}),xs=_o(function(e,t){var n=Z(t,Tr(xs));return fr(e,ke,void 0,t,n)}),js=_o(function(e,t){var n=Z(t,Tr(js));return fr(e,xe,void 0,t,n)}),vs=jr(function(e,t){return fr(e,ve,void 0,void 0,void 0,t)}),Ss=cr(gn),As=cr(function(e,t){return e>=t}),Ts=On(function(){return arguments}())?On:function(e){return Bi(e)&&Id.call(e,"callee")&&!Md.call(e,"callee")},zs=gd.isArray,Is=En?W(En):function(e){return Bi(e)&&hn(e)==Ge},Os=ea||hd,Es=Fn?W(Fn):function(e){return Bi(e)&&hn(e)==We},Fs=Nn?W(Nn):function(e){return Bi(e)&&Pa(e)==Be},Ns=Rn?W(Rn):function(e){return Bi(e)&&hn(e)==De},Rs=Wn?W(Wn):function(e){return Bi(e)&&Pa(e)==qe},Ws=Cn?W(Cn):function(e){return Bi(e)&&Li(e.length)&&!!mn[hn(e)]},Cs=cr(Gn),Ls=cr(function(e,t){return e<=t}),Ps=Ho(function(e,t){if(Kr(t)||Fi(t))return void $o(t,Xi(t),e);for(var n in t)Id.call(t,n)&&Mt(e,n,t[n])}),Bs=Ho(function(e,t){$o(t,ed(t),e)}),Us=Ho(function(e,t,n,o){$o(t,ed(t),e,o)}),Vs=Ho(function(e,t,n,o){$o(t,Xi(t),e,o)}),Ms=jr(Gt),Ds=_o(function(e,t){e=bd(e);var n=-1,o=t.length,r=2<o?t[2]:void 0;for(r&&Mr(t[0],t[1],r)&&(o=1);++n<o;)for(var i=t[n],d=ed(i),a=-1,s=d.length;++a<s;){var l=d[a],p=e[l];(void 0===p||Ei(p,Ad[l])&&!Id.call(e,l))&&(e[l]=i[l])}return e}),qs=_o(function(e){return e.push(void 0,mr),r(Hs,void 0,e)}),$s=dr(function(e,t,n){null!=t&&"function"!=typeof t.toString&&(t=Fd.call(t)),e[t]=n},ad(sd)),Js=dr(function(e,t,n){null!=t&&"function"!=typeof t.toString&&(t=Fd.call(t)),Id.call(e,t)?e[t].push(n):e[t]=[n]},zr),Ks=_o(In),Gs=Ho(function(e,t,n){Qn(e,t,n)}),Hs=Ho(function(e,t,n,o){Qn(e,t,n,o)}),Zs=jr(function(e,t){var n={};if(null==e)return n;var o=!1;t=u(t,function(t){return t=Fo(t,e),o||(o=1<t.length),t}),$o(e,Sr(e),n),o&&(n=Zt(n,_e|ce|ue,wr));for(var r=t.length;r--;)vo(n,t[r]);return n}),Ys=jr(function(e,t){return null==e?{}:no(e,t)}),Qs=gr(Xi),Xs=gr(ed),el=er(function(e,t,n){return t=t.toLowerCase(),e+(n?rd(t):t)}),tl=er(function(e,t,n){return e+(n?"-":"")+t.toLowerCase()}),nl=er(function(e,t,n){return e+(n?" ":"")+t.toLowerCase()}),ol=Xo("toLowerCase"),rl=er(function(e,t,n){return e+(n?"_":"")+t.toLowerCase()}),il=er(function(e,t,n){return e+(n?" ":"")+al(t)}),dl=er(function(e,t,n){return e+(n?" ":"")+t.toUpperCase()}),al=Xo("toUpperCase"),sl=_o(function(e,t){try{return r(e,void 0,t)}catch(t){return Ri(t)?t:new yd(t)}}),ll=jr(function(e,t){return d(t,function(t){t=di(t),Kt(e,t,ys(e[t],e))}),e}),pl=rr(),_l=rr(!0),cl=_o(function(e,t){return function(n){return In(n,e,t)}}),ul=_o(function(e,t){return function(n){return In(e,n,t)}}),hl=sr(u),gl=sr(s),fl=sr(y),yl=_r(),ml=_r(!0),wl=ar(function(e,t){return e+t},0),bl=hr("ceil"),kl=ar(function(e,t){return e/t},1),xl=hr("floor"),jl=ar(function(e,t){return e*t},1),vl=hr("round"),Sl=ar(function(e,t){return e-t},0);return o.after=function(e,t){if("function"!=typeof t)throw new jd(se);return e=Ji(e),function(){if(1>--e)return t.apply(this,arguments)}},o.ary=vi,o.assign=Ps,o.assignIn=Bs,o.assignInWith=Us,o.assignWith=Vs,o.at=Ms,o.before=Si,o.bind=ys,o.bindAll=ll,o.bindKey=ms,o.castArray=function(){if(!arguments.length)return[];var e=arguments[0];return zs(e)?e:[e]},o.chain=wi,o.chunk=function(e,t,n){t=(n?Mr(e,t,n):void 0===t)?1:ra(Ji(t),0);var o=null==e?0:e.length;if(!o||1>t)return[];for(var r=0,i=0,d=gd(Yd(o/t));r<o;)d[i++]=fo(e,r,r+=t);return d},o.compact=function(e){for(var t=-1,n=null==e?0:e.length,o=0,r=[],i;++t<n;)i=e[t],i&&(r[o++]=i);return r},o.concat=function(){var e=arguments.length;if(!e)return[];for(var t=gd(e-1),n=arguments[0],o=e;o--;)t[o-1]=arguments[o];return h(zs(n)?qo(n):[n],dn(t,1))},o.cond=function(e){var t=null==e?0:e.length,n=zr();return e=t?u(e,function(e){if("function"!=typeof e[1])throw new jd(se);return[n(e[0]),e[1]]}):[],_o(function(n){for(var o=-1,i;++o<t;)if(i=e[o],r(i[0],this,n))return r(i[1],this,n)})},o.conforms=function(e){return Yt(Zt(e,_e))},o.constant=ad,o.countBy=ss,o.create=function(e,t){var n=Aa(e);return null==t?n:$t(n,t)},o.curry=Ai,o.curryRight=Ti,o.debounce=zi,o.defaults=Ds,o.defaultsDeep=qs,o.defer=ws,o.delay=bs,o.difference=qa,o.differenceBy=$a,o.differenceWith=Ja,o.drop=function(e,t,n){var o=null==e?0:e.length;return o?(t=n||void 0===t?1:Ji(t),fo(e,0>t?0:t,o)):[]},o.dropRight=function(e,t,n){var o=null==e?0:e.length;return o?(t=n||void 0===t?1:Ji(t),t=o-t,fo(e,0,0>t?0:t)):[]},o.dropRightWhile=function(e,t){return e&&e.length?Ao(e,zr(t,3),!0,
577!0):[]},o.dropWhile=function(e,t){return e&&e.length?Ao(e,zr(t,3),!0):[]},o.fill=function(e,t,n,o){var r=null==e?0:e.length;return r?(n&&"number"!=typeof n&&Mr(e,t,n)&&(n=0,o=r),on(e,t,n,o)):[]},o.filter=function(e,t){var n=zs(e)?l:rn;return n(e,zr(t,3))},o.flatMap=function(e,t){return dn(ji(e,t),1)},o.flatMapDeep=function(e,t){return dn(ji(e,t),Te)},o.flatMapDepth=function(e,t,n){return n=void 0===n?1:Ji(n),dn(ji(e,t),n)},o.flatten=ci,o.flattenDeep=function(e){var t=null==e?0:e.length;return t?dn(e,Te):[]},o.flattenDepth=function(e,t){var n=null==e?0:e.length;return n?(t=void 0===t?1:Ji(t),dn(e,t)):[]},o.flip=function(e){return fr(e,Se)},o.flow=pl,o.flowRight=_l,o.fromPairs=function(e){for(var t=-1,n=null==e?0:e.length,o={},r;++t<n;)r=e[t],o[r[0]]=r[1];return o},o.functions=function(e){return null==e?[]:ln(e,Xi(e))},o.functionsIn=function(e){return null==e?[]:ln(e,ed(e))},o.groupBy=_s,o.initial=function(e){var t=null==e?0:e.length;return t?fo(e,0,-1):[]},o.intersection=Ka,o.intersectionBy=Ga,o.intersectionWith=Ha,o.invert=$s,o.invertBy=Js,o.invokeMap=cs,o.iteratee=ld,o.keyBy=us,o.keys=Xi,o.keysIn=ed,o.map=ji,o.mapKeys=function(e,t){var n={};return t=zr(t,3),an(e,function(e,o,r){Kt(n,t(e,o,r),e)}),n},o.mapValues=function(e,t){var n={};return t=zr(t,3),an(e,function(e,o,r){Kt(n,o,t(e,o,r))}),n},o.matches=function(e){return Zn(Zt(e,_e))},o.matchesProperty=function(e,t){return Yn(e,Zt(t,_e))},o.memoize=Ii,o.merge=Gs,o.mergeWith=Hs,o.method=cl,o.methodOf=ul,o.mixin=pd,o.negate=Oi,o.nthArg=function(e){return e=Ji(e),_o(function(t){return eo(t,e)})},o.omit=Zs,o.omitBy=function(e,t){return td(e,Oi(zr(t)))},o.once=function(e){return Si(2,e)},o.orderBy=function(e,t,n,o){return null==e?[]:(zs(t)||(t=null==t?[]:[t]),n=o?void 0:n,zs(n)||(n=null==n?[]:[n]),to(e,t,n))},o.over=hl,o.overArgs=ks,o.overEvery=gl,o.overSome=fl,o.partial=xs,o.partialRight=js,o.partition=hs,o.pick=Ys,o.pickBy=td,o.property=cd,o.propertyOf=function(e){return function(t){return null==e?void 0:cn(e,t)}},o.pull=Za,o.pullAll=gi,o.pullAllBy=function(e,t,n){return e&&e.length&&t&&t.length?io(e,t,zr(n,2)):e},o.pullAllWith=function(e,t,n){return e&&e.length&&t&&t.length?io(e,t,void 0,n):e},o.pullAt=Ya,o.range=yl,o.rangeRight=ml,o.rearg=vs,o.reject=function(e,t){var n=zs(e)?l:rn;return n(e,Oi(zr(t,3)))},o.remove=function(e,t){var n=[];if(!(e&&e.length))return n;var o=-1,r=[],i=e.length;for(t=zr(t,3);++o<i;){var d=e[o];t(d,o,e)&&(n.push(d),r.push(o))}return ao(e,r),n},o.rest=function(e,t){if("function"!=typeof e)throw new jd(se);return t=void 0===t?t:Ji(t),_o(e,t)},o.reverse=fi,o.sampleSize=function(e,t,n){t=(n?Mr(e,t,n):void 0===t)?1:Ji(t);var o=zs(e)?Bt:uo;return o(e,t)},o.set=function(e,t,n){return null==e?e:ho(e,t,n)},o.setWith=function(e,t,n,o){return o="function"==typeof o?o:void 0,null==e?e:ho(e,t,n,o)},o.shuffle=function(e){var t=zs(e)?Ut:go;return t(e)},o.slice=function(e,t,n){var o=null==e?0:e.length;return o?(n&&"number"!=typeof n&&Mr(e,t,n)?(t=0,n=o):(t=null==t?0:Ji(t),n=void 0===n?o:Ji(n)),fo(e,t,n)):[]},o.sortBy=gs,o.sortedUniq=function(e){return e&&e.length?bo(e):[]},o.sortedUniqBy=function(e,t){return e&&e.length?bo(e,zr(t,2)):[]},o.split=function(e,t,n){return(n&&"number"!=typeof n&&Mr(e,t,n)&&(t=n=void 0),n=void 0===n?Oe:n>>>0,!n)?[]:(e=Zi(e),e&&("string"==typeof t||null!=t&&!Ns(t))&&(t=xo(t),!t&&q(e))?No(ne(e),0,n):e.split(t,n))},o.spread=function(e,t){if("function"!=typeof e)throw new jd(se);return t=null==t?0:ra(Ji(t),0),_o(function(n){var o=n[t],i=No(n,0,t);return o&&h(i,o),r(e,this,i)})},o.tail=function(e){var t=null==e?0:e.length;return t?fo(e,1,t):[]},o.take=function(e,t,n){return e&&e.length?(t=n||void 0===t?1:Ji(t),fo(e,0,0>t?0:t)):[]},o.takeRight=function(e,t,n){var o=null==e?0:e.length;return o?(t=n||void 0===t?1:Ji(t),t=o-t,fo(e,0>t?0:t,o)):[]},o.takeRightWhile=function(e,t){return e&&e.length?Ao(e,zr(t,3),!1,!0):[]},o.takeWhile=function(e,t){return e&&e.length?Ao(e,zr(t,3)):[]},o.tap=function(e,t){return t(e),e},o.throttle=function(e,t,n){var o=!0,r=!0;if("function"!=typeof e)throw new jd(se);return Pi(n)&&(o="leading"in n?!!n.leading:o,r="trailing"in n?!!n.trailing:r),zi(e,t,{leading:o,maxWait:t,trailing:r})},o.thru=bi,o.toArray=qi,o.toPairs=Qs,o.toPairsIn=Xs,o.toPath=function(e){return zs(e)?u(e,di):Di(e)?[e]:qo(Da(Zi(e)))},o.toPlainObject=Hi,o.transform=function(e,t,n){var o=zs(e),r=o||Os(e)||Ws(e);if(t=zr(t,4),null==n){var i=e&&e.constructor;n=r?o?new i:[]:Pi(e)?Wi(i)?Aa(Ud(e)):{}:{}}return(r?d:an)(e,function(e,o,r){return t(n,e,o,r)}),n},o.unary=function(e){return vi(e,1)},o.union=Qa,o.unionBy=Xa,o.unionWith=es,o.uniq=function(e){return e&&e.length?jo(e):[]},o.uniqBy=function(e,t){return e&&e.length?jo(e,zr(t,2)):[]},o.uniqWith=function(e,t){return t="function"==typeof t?t:void 0,e&&e.length?jo(e,void 0,t):[]},o.unset=function(e,t){return null==e||vo(e,t)},o.unzip=yi,o.unzipWith=mi,o.update=function(e,t,n){return null==e?e:So(e,t,Eo(n))},o.updateWith=function(e,t,n,o){return o="function"==typeof o?o:void 0,null==e?e:So(e,t,Eo(n),o)},o.values=od,o.valuesIn=function(e){return null==e?[]:C(e,ed(e))},o.without=ts,o.words=dd,o.wrap=function(e,t){return xs(Eo(t),e)},o.xor=ns,o.xorBy=os,o.xorWith=rs,o.zip=is,o.zipObject=function(e,t){return Io(e||[],t||[],Mt)},o.zipObjectDeep=function(e,t){return Io(e||[],t||[],ho)},o.zipWith=ds,o.entries=Qs,o.entriesIn=Xs,o.extend=Bs,o.extendWith=Us,pd(o,o),o.add=wl,o.attempt=sl,o.camelCase=el,o.capitalize=rd,o.ceil=bl,o.clamp=function(e,t,n){return void 0===n&&(n=t,t=void 0),void 0!==n&&(n=Gi(n),n=n===n?n:0),void 0!==t&&(t=Gi(t),t=t===t?t:0),Ht(Gi(e),t,n)},o.clone=function(e){return Zt(e,ue)},o.cloneDeep=function(e){return Zt(e,_e|ue)},o.cloneDeepWith=function(e,t){return t="function"==typeof t?t:void 0,Zt(e,_e|ue,t)},o.cloneWith=function(e,t){return t="function"==typeof t?t:void 0,Zt(e,ue,t)},o.conformsTo=function(e,t){return null==t||Qt(e,t,Xi(t))},o.deburr=id,o.defaultTo=function(e,t){return null==e||e!==e?t:e},o.divide=kl,o.endsWith=function(e,t,n){e=Zi(e),t=xo(t);var o=e.length;n=void 0===n?o:Ht(Ji(n),0,o);var r=n;return n-=t.length,0<=n&&e.slice(n,r)==t},o.eq=Ei,o.escape=function(e){return e=Zi(e),e&&_t.test(e)?e.replace(lt,Bn):e},o.escapeRegExp=function(e){return e=Zi(e),e&&wt.test(e)?e.replace(mt,"\\$&"):e},o.every=function(e,t,n){var o=zs(e)?s:tn;return n&&Mr(e,t,n)&&(t=void 0),o(e,zr(t,3))},o.find=ls,o.findIndex=pi,o.findKey=function(e,t){return b(e,zr(t,3),an)},o.findLast=ps,o.findLastIndex=_i,o.findLastKey=function(e,t){return b(e,zr(t,3),sn)},o.floor=xl,o.forEach=ki,o.forEachRight=xi,o.forIn=function(e,t){return null==e?e:Ia(e,zr(t,3),ed)},o.forInRight=function(e,t){return null==e?e:Oa(e,zr(t,3),ed)},o.forOwn=function(e,t){return e&&an(e,zr(t,3))},o.forOwnRight=function(e,t){return e&&sn(e,zr(t,3))},o.get=Yi,o.gt=Ss,o.gte=As,o.has=function(e,t){return null!=e&&Wr(e,t,bn)},o.hasIn=Qi,o.head=ui,o.identity=sd,o.includes=function(e,t,n,o){e=Fi(e)?e:od(e),n=n&&!o?Ji(n):0;var r=e.length;return 0>
577n&&(n=ra(r+n,0)),Mi(e)?n<=r&&-1<e.indexOf(t,n):!!r&&-1<x(e,t,n)},o.indexOf=function(e,t,n){var o=null==e?0:e.length;if(!o)return-1;var r=null==n?0:Ji(n);return 0>r&&(r=ra(o+r,0)),x(e,t,r)},o.inRange=function(e,t,n){return t=$i(t),void 0===n?(n=t,t=0):n=$i(n),e=Gi(e),vn(e,t,n)},o.invoke=Ks,o.isArguments=Ts,o.isArray=zs,o.isArrayBuffer=Is,o.isArrayLike=Fi,o.isArrayLikeObject=Ni,o.isBoolean=function(e){return!0===e||!1===e||Bi(e)&&hn(e)==Re},o.isBuffer=Os,o.isDate=Es,o.isElement=function(e){return Bi(e)&&1===e.nodeType&&!Vi(e)},o.isEmpty=function(e){if(null==e)return!0;if(Fi(e)&&(zs(e)||"string"==typeof e||"function"==typeof e.splice||Os(e)||Ws(e)||Ts(e)))return!e.length;var t=Pa(e);if(t==Be||t==qe)return!e.size;if(Kr(e))return!Jn(e).length;for(var n in e)if(Id.call(e,n))return!1;return!0},o.isEqual=function(e,t){return Ln(e,t)},o.isEqualWith=function(e,t,n){n="function"==typeof n?n:void 0;var o=n?n(e,t):void 0;return void 0===o?Ln(e,t,void 0,n):!!o},o.isError=Ri,o.isFinite=function(e){return"number"==typeof e&&ta(e)},o.isFunction=Wi,o.isInteger=Ci,o.isLength=Li,o.isMap=Fs,o.isMatch=function(e,t){return e===t||Dn(e,t,Or(t))},o.isMatchWith=function(e,t,n){return n="function"==typeof n?n:void 0,Dn(e,t,Or(t),n)},o.isNaN=function(e){return Ui(e)&&e!=+e},o.isNative=function(e){if(Ba(e))throw new yd("Unsupported core-js use. Try https://npms.io/search?q=ponyfill.");return qn(e)},o.isNil=function(e){return null==e},o.isNull=function(e){return null===e},o.isNumber=Ui,o.isObject=Pi,o.isObjectLike=Bi,o.isPlainObject=Vi,o.isRegExp=Ns,o.isSafeInteger=function(e){return Ci(e)&&e>=-ze&&e<=ze},o.isSet=Rs,o.isString=Mi,o.isSymbol=Di,o.isTypedArray=Ws,o.isUndefined=function(e){return void 0===e},o.isWeakMap=function(e){return Bi(e)&&Pa(e)==Ke},o.isWeakSet=function(e){return Bi(e)&&hn(e)=="[object WeakSet]"},o.join=function(e,t){return null==e?"":na.call(e,t)},o.kebabCase=tl,o.last=hi,o.lastIndexOf=function(e,t,n){var o=null==e?0:e.length;if(!o)return-1;var r=o;return void 0!==n&&(r=Ji(n),r=0>r?ra(o+r,0):ia(r,o-1)),t===t?ee(e,t,r):k(e,v,r,!0)},o.lowerCase=nl,o.lowerFirst=ol,o.lt=Cs,o.lte=Ls,o.max=function(e){return e&&e.length?nn(e,sd,gn):void 0},o.maxBy=function(e,t){return e&&e.length?nn(e,zr(t,2),gn):void 0},o.mean=function(e){return S(e,sd)},o.meanBy=function(e,t){return S(e,zr(t,2))},o.min=function(e){return e&&e.length?nn(e,sd,Gn):void 0},o.minBy=function(e,t){return e&&e.length?nn(e,zr(t,2),Gn):void 0},o.stubArray=ud,o.stubFalse=hd,o.stubObject=function(){return{}},o.stubString=function(){return""},o.stubTrue=function(){return!0},o.multiply=jl,o.nth=function(e,t){return e&&e.length?eo(e,Ji(t)):void 0},o.noConflict=function(){return Sn._===this&&(Sn._=Rd),this},o.noop=_d,o.now=fs,o.pad=function(e,t,n){e=Zi(e),t=Ji(t);var o=t?te(e):0;if(!t||o>=t)return e;var r=(t-o)/2;return lr(Qd(r),n)+e+lr(Yd(r),n)},o.padEnd=function(e,t,n){e=Zi(e),t=Ji(t);var o=t?te(e):0;return t&&o<t?e+lr(t-o,n):e},o.padStart=function(e,t,n){e=Zi(e),t=Ji(t);var o=t?te(e):0;return t&&o<t?lr(t-o,n)+e:e},o.parseInt=function(e,t,n){return n||null==t?t=0:t&&(t=+t),aa(Zi(e).replace(bt,""),t||0)},o.random=function(e,t,n){if(n&&"boolean"!=typeof n&&Mr(e,t,n)&&(t=n=void 0),void 0===n&&("boolean"==typeof t?(n=t,t=void 0):"boolean"==typeof e&&(n=e,e=void 0)),void 0===e&&void 0===t?(e=0,t=1):(e=$i(e),void 0===t?(t=e,e=0):t=$i(t)),e>t){var o=e;e=t,t=o}if(n||e%1||t%1){var r=sa();return ia(e+r*(t-e+kn("1e-"+((r+"").length-1))),t)}return so(e,t)},o.reduce=function(e,t,n){var o=zs(e)?g:I,r=3>arguments.length;return o(e,zr(t,4),n,r,Ta)},o.reduceRight=function(e,t,n){var o=zs(e)?f:I,r=3>arguments.length;return o(e,zr(t,4),n,r,za)},o.repeat=function(e,t,n){return t=(n?Mr(e,t,n):void 0===t)?1:Ji(t),po(Zi(e),t)},o.replace=function(){var e=arguments,t=Zi(e[0]);return 3>e.length?t:t.replace(e[1],e[2])},o.result=nd,o.round=vl,o.runInContext=t,o.sample=function(e){var t=zs(e)?Pt:co;return t(e)},o.size=function(e){if(null==e)return 0;if(Fi(e))return Mi(e)?te(e):e.length;var t=Pa(e);return t==Be||t==qe?e.size:Jn(e).length},o.snakeCase=rl,o.some=function(e,t,n){var o=zs(e)?y:yo;return n&&Mr(e,t,n)&&(t=void 0),o(e,zr(t,3))},o.sortedIndex=function(e,t){return mo(e,t)},o.sortedIndexBy=function(e,t,n){return wo(e,t,zr(n,2))},o.sortedIndexOf=function(e,t){var n=null==e?0:e.length;if(n){var o=mo(e,t);if(o<n&&Ei(e[o],t))return o}return-1},o.sortedLastIndex=function(e,t){return mo(e,t,!0)},o.sortedLastIndexBy=function(e,t,n){return wo(e,t,zr(n,2),!0)},o.sortedLastIndexOf=function(e,t){var n=null==e?0:e.length;if(n){var o=mo(e,t,!0)-1;if(Ei(e[o],t))return o}return-1},o.startCase=il,o.startsWith=function(e,t,n){return e=Zi(e),n=null==n?0:Ht(Ji(n),0,e.length),t=xo(t),e.slice(n,n+t.length)==t},o.subtract=Sl,o.sum=function(e){return e&&e.length?E(e,sd):0},o.sumBy=function(e,t){return e&&e.length?E(e,zr(t,2)):0},o.template=function(e,t,n){var r=o.templateSettings;n&&Mr(e,t,n)&&(t=void 0),e=Zi(e),t=Us({},t,r,yr);var i=Us({}
577,t.imports,r.imports,yr),d=Xi(i),a=C(i,d),s=0,l=t.interpolate||Ct,p="__p += '",_=kd((t.escape||Ct).source+"|"+l.source+"|"+(l===ht?zt:Ct).source+"|"+(t.evaluate||Ct).source+"|$","g"),c="//# sourceURL="+(Id.call(t,"sourceURL")?(t.sourceURL+"").replace(/\s/g," "):"lodash.templateSources["+ ++yn+"]")+"\n",u,h;e.replace(_,function(t,n,o,r,i,d){return o||(o=r),p+=e.slice(s,d).replace(Lt,M),n&&(u=!0,p+="' +\n__e("+n+") +\n'"),i&&(h=!0,p+="';\n"+i+";\n__p += '"),o&&(p+="' +\n((__t = ("+o+")) == null ? '' : __t) +\n'"),s=d+t.length,t}),p+="';\n";var g=Id.call(t,"variable")&&t.variable;if(!g)p="with (obj) {\n"+p+"\n}\n";else if(At.test(g))throw new yd("Invalid `variable` option passed into `_.template`");p=(h?p.replace(it,""):p).replace(dt,"$1").replace(at,"$1;"),p="function("+(g||"obj")+") {\n"+(g?"":"obj || (obj = {});\n")+"var __t, __p = ''"+(u?", __e = _.escape":"")+(h?", __j = Array.prototype.join;\nfunction print() { __p += __j.call(arguments, '') }\n":";\n")+p+"return __p\n}";var f=sl(function(){return md(d,c+"return "+p).apply(void 0,a)});if(f.source=p,Ri(f))throw f;return f},o.times=function(e,t){if(e=Ji(e),1>e||e>ze)return[];var n=Oe,o=ia(e,Oe);t=zr(t),e-=Oe;for(var r=F(o,t);++n<e;)t(n);return r},o.toFinite=$i,o.toInteger=Ji,o.toLength=Ki,o.toLower=function(e){return Zi(e).toLowerCase()},o.toNumber=Gi,o.toSafeInteger=function(e){return e?Ht(Ji(e),-ze,ze):0===e?e:0},o.toString=Zi,o.toUpper=function(e){return Zi(e).toUpperCase()},o.trim=function(e,t,n){if(e=Zi(e),e&&(n||void 0===t))return R(e);if(!e||!(t=xo(t)))return e;var o=ne(e),r=ne(t),i=P(o,r),d=B(o,r)+1;return No(o,i,d).join("")},o.trimEnd=function(e,t,n){if(e=Zi(e),e&&(n||void 0===t))return e.slice(0,oe(e)+1);if(!e||!(t=xo(t)))return e;var o=ne(e),r=B(o,ne(t))+1;return No(o,0,r).join("")},o.trimStart=function(e,t,n){if(e=Zi(e),e&&(n||void 0===t))return e.replace(bt,"");if(!e||!(t=xo(t)))return e;var o=ne(e),r=P(o,ne(t));return No(o,r).join("")},o.truncate=function(e,t){var n=30,o="...";if(Pi(t)){var r="separator"in t?t.separator:r;n="length"in t?Ji(t.length):n,o="omission"in t?xo(t.omission):o}e=Zi(e);var i=e.length;if(q(e)){var d=ne(e);i=d.length}if(n>=i)return e;var a=n-te(o);if(1>a)return o;var s=d?No(d,0,a).join(""):e.slice(0,a);if(void 0===r)return s+o;if(d&&(a+=s.length-a),Ns(r)){if(e.slice(a).search(r)){var l=s,p;for(r.global||(r=kd(r.source,Zi(It.exec(r))+"g")),r.lastIndex=0;p=r.exec(l);)var _=p.index;s=s.slice(0,void 0===_?a:_)}}else if(e.indexOf(xo(r),a)!=a){var c=s.lastIndexOf(r);-1<c&&(s=s.slice(0,c))}return s+o},o.unescape=function(e){return e=Zi(e),e&&pt.test(e)?e.replace(st,Un):e},o.uniqueId=function(e){var t=++Od;return Zi(e)+t},o.upperCase=dl,o.upperFirst=al,o.each=ki,o.eachRight=xi,o.first=ui,pd(o,function(){var e={};return an(o,function(t,n){Id.call(o.prototype,n)||(e[n]=t)}),e}(),{chain:!1}),o.VERSION="4.17.21",d(["bind","bindKey","curry","curryRight","partial","partialRight"],function(e){o[e].placeholder=o}),d(["drop","take"],function(e,t){z.prototype[e]=function(o){o=void 0===o?1:ra(Ji(o),0);var n=this.__filtered__&&!t?new z(this):this.clone();return n.__filtered__?n.__takeCount__=ia(o,n.__takeCount__):n.__views__.push({size:ia(o,Oe),type:e+(0>n.__dir__?"Right":"")}),n},z.prototype[e+"Right"]=function(t){return this.reverse()[e](t).reverse()}}),d(["filter","map","takeWhile"],function(e,t){var n=t+1;z.prototype[e]=function(e){var t=this.clone();return t.__iteratees__.push({iteratee:zr(e,3),type:n}),t.__filtered__=t.__filtered__||n==Ae||n==3,t}}),d(["head","last"],function(e,t){var n="take"+(t?"Right":"");z.prototype[e]=function(){return this[n](1).value()[0]}}),d(["initial","tail"],function(e,t){var n="drop"+(t?"":"Right");z.prototype[e]=function(){return this.__filtered__?new z(this):this[n](1)}}),z.prototype.compact=function(){return this.filter(sd)},z.prototype.find=function(e){return this.filter(e).head()},z.prototype.findLast=function(e){return this.reverse().find(e)},z.prototype.invokeMap=_o(function(e,t){return"function"==typeof e?new z(this):this.map(function(n){return In(n,e,t)})}),z.prototype.reject=function(e){return this.filter(Oi(zr(e)))},z.prototype.slice=function(e,t){e=Ji(e);var n=this;return n.__filtered__&&(0<e||0>t)?new z(n):(0>e?n=n.takeRight(-e):e&&(n=n.drop(e)),void 0!==t&&(t=Ji(t),n=0>t?n.dropRight(-t):n.take(t-e)),n)},z.prototype.takeRightWhile=function(e){return this.reverse().takeWhile(e).reverse()},z.prototype.toArray=function(){return this.take(Oe)},an(z.prototype,function(e,t){var n=/^(?:filter|find|map|reject)|While$/.test(t),r=/^(?:head|last)$/.test(t),i=o[r?"take"+("last"==t?"Right":""):t],d=r||/^find/.test(t);i&&(o.prototype[t]=function(){var t=this.__wrapped__,a=r?[1]:arguments,s=t instanceof z,l=a[0],p=s||zs(t),_=function(e){var t=i.apply(o,h([e],a));return r&&c?t[0]:t};p&&n&&"function"==typeof l&&1!=l.length&&(s=p=!1);var c=this.__chain__,u=!!this.__actions__.length,g=d&&!c,f=s&&!u;if(!d&&p){t=f?t:new z(this);var y=e.apply(t,a);return y.__actions__.push({func:bi,args:[_],thisArg:void 0}),new m(y,c)}
577return g&&f?e.apply(this,a):(y=this.thru(_),g?r?y.value()[0]:y.value():y)})}),d(["pop","push","shift","sort","splice","unshift"],function(e){var t=vd[e],n=/^(?:push|sort|unshift)$/.test(e)?"tap":"thru",r=/^(?:pop|shift)$/.test(e);o.prototype[e]=function(){var e=arguments;if(r&&!this.__chain__){var o=this.value();return t.apply(zs(o)?o:[],e)}return this[n](function(n){return t.apply(zs(n)?n:[],e)})}}),an(z.prototype,function(e,t){var n=o[t];if(n){var r=n.name+"";Id.call(ya,r)||(ya[r]=[]),ya[r].push({name:t,func:n})}}),ya[ir(void 0,ye).name]=[{name:"wrapper",func:void 0}],z.prototype.clone=function(){var e=new z(this.__wrapped__);return e.__actions__=qo(this.__actions__),e.__dir__=this.__dir__,e.__filtered__=this.__filtered__,e.__iteratees__=qo(this.__iteratees__),e.__takeCount__=this.__takeCount__,e.__views__=qo(this.__views__),e},z.prototype.reverse=function(){if(this.__filtered__){var e=new z(this);e.__dir__=-1,e.__filtered__=!0}else e=this.clone(),e.__dir__*=-1;return e},z.prototype.value=function(){var e=this.__wrapped__.value(),t=this.__dir__,n=zs(e),o=0>t,r=n?e.length:0,i=Nr(0,r,this.__views__),d=i.start,a=i.end,s=a-d,l=o?a:d-1,p=this.__iteratees__,_=p.length,c=0,u=ia(s,this.__takeCount__);if(!n||!o&&r==s&&u==s)return To(e,this.__actions__);var h=[];outer:for(;s--&&c<u;){l+=t;for(var g=-1,f=e[l];++g<_;){var y=p[g],m=y.iteratee,w=y.type,b=m(f);if(w==2)f=b;else if(!b)if(w==Ae)continue outer;else break outer}h[c++]=f}return h},o.prototype.at=as,o.prototype.chain=function(){return wi(this)},o.prototype.commit=function(){return new m(this.value(),this.__chain__)},o.prototype.next=function(){void 0===this.__values__&&(this.__values__=qi(this.value()));var e=this.__index__>=this.__values__.length,t=e?void 0:this.__values__[this.__index__++];return{done:e,value:t}},o.prototype.plant=function(e){for(var t=this,n,o;t instanceof _;){o=li(t),o.__index__=0,o.__values__=void 0,n?r.__wrapped__=o:n=o;var r=o;t=t.__wrapped__}return r.__wrapped__=e,n},o.prototype.reverse=function(){var e=this.__wrapped__;if(e instanceof z){var t=e;return this.__actions__.length&&(t=new z(this)),t=t.reverse(),t.__actions__.push({func:bi,args:[fi],thisArg:void 0}),new m(t,this.__chain__)}return this.thru(fi)},o.prototype.toJSON=o.prototype.valueOf=o.prototype.value=function(){return To(this.__wrapped__,this.__actions__)},o.prototype.first=o.prototype.head,$d&&(o.prototype[$d]=function(){return this}),o}();"function"==typeof G&&"object"==typeof G.amd&&G.amd?(Sn._=Vn,G(function(){return Vn})):Tn?((Tn.exports=Vn)._=Vn,An._=Vn):Sn._=Vn}).call(this)}).call(this)}).call(this,"undefined"==typeof global?"undefined"==typeof self?"undefined"==typeof window?{}:window:self:global)},{}],4:[function(t,n,o){(function(){var t=Math.log,r=function(e){var t=new r.Index;return t.pipeline.add(r.trimmer,r.stopWordFilter,r.stemmer),e&&e.call(t,t),t};r.version="1.0.0",r.utils={},r.utils.warn=function(e){return function(t){e.console&&console.warn&&console.warn(t)}}(this),r.utils.asString=function(e){return void 0===e||null===e?"":e.toString()},r.EventEmitter=function(){this.events={}},r.EventEmitter.prototype.addListener=function(){var e=Array.prototype.slice.call(arguments),t=e.pop();if("function"!=typeof t)throw new TypeError("last argument must be a function");e.forEach(function(e){this.hasHandler(e)||(this.events[e]=[]),this.events[e].push(t)},this)},r.EventEmitter.prototype.removeListener=function(e,t){if(this.hasHandler(e)){var n=this.events[e].indexOf(t);this.events[e].splice(n,1),this.events[e].length||delete this.events[e]}},r.EventEmitter.prototype.emit=function(e){if(this.hasHandler(e)){var t=Array.prototype.slice.call(arguments,1);this.events[e].forEach(function(e){e.apply(void 0,t)})}},r.EventEmitter.prototype.hasHandler=function(e){return e in this.events},r.tokenizer=function(e){return arguments.length&&null!=e&&void 0!=e?Array.isArray(e)?e.map(function(e){return r.utils.asString(e).toLowerCase()}):e.toString().trim().toLowerCase().split(r.tokenizer.separator):[]},r.tokenizer.separator=/[\s\-]+/,r.tokenizer.load=function(e){var t=this.registeredFunctions[e];if(!t)throw new Error("Cannot load un-registered function: "+e);return t},r.tokenizer.label="default",r.tokenizer.registeredFunctions={default:r.tokenizer},r.tokenizer.registerFunction=function(e,t){t in this.registeredFunctions&&r.utils.warn("Overwriting existing tokenizer: "+t),e.label=t,this.registeredFunctions[t]=e},r.Pipeline=function(){this._stack=[]},r.Pipeline.registeredFunctions={},r.Pipeline.registerFunction=function(e,t){t in this.registeredFunctions&&r.utils.warn("Overwriting existing registered function: "+t),e.label=t,r.Pipeline.registeredFunctions[e.label]=e},r.Pipeline.warnIfFunctionNotRegistered=function(e){var t=e.label&&e.label in this.registeredFunctions;t||r.utils.warn("Function is not registered with pipeline. This may cause problems when serialising the index.\n",e)},r.Pipeline.load=function(e){var t=new r.Pipeline;return e.forEach(function(e){var n=r.Pipeline.registeredFunctions[e];if(n)t.add(n);else throw new Error("Cannot load un-registered function: "+e)}),t},r.Pipeline.prototype.add=function(){var e=Array.prototype.slice.call(arguments);e.forEach(function(e){r.Pipeline.warnIfFunctionNotRegistered(e),this._stack.push(e)},this)},r.Pipeline.prototype.after=function(e,t){r.Pipeline.warnIfFunctionNotRegistered(t);var n=this._stack.indexOf(e);if(-1==n)throw new Error("Cannot find existingFn");++n,this._stack.splice(n,0,t)},r.Pipeline.prototype.before=function(e,t){r.Pipeline.warnIfFunctionNotRegistered(t);var n=this._stack.indexOf(e);if(-1==n)throw new Error("Cannot find existingFn");this._stack.splice(n,0,t)},r.Pipeline.prototype.remove=function(e){var t=this._stack.indexOf(e);-1==t||this._stack.splice(t,1)},r.Pipeline.prototype.run=function(e){for(var t=[],n=e.length,o=this._stack.length,r=0,i;r<n;r++){i=e[r];for(var d=0;d<o&&(i=this._stack[d](i,r,e),void 0!==i&&""!==i);d++);void 0!==i&&""!==i&&t.push(i)}return t},r.Pipeline.prototype.reset=function(){this._stack=[]},r.Pipeline.prototype.toJSON=function(){return this._stack.map(function(e){return r.Pipeline.warnIfFunctionNotRegistered(e),e.label})},r.Vector=function(){this._magnitude=null,this.list=void 0,this.length=0},r.Vector.Node=function(e,t,n){this.idx=e,this.val=t,this.next=n},r.Vector.prototype.insert=function(e,t){this._magnitude=void 0;var n=this.list;if(!n)return this.list=new r.Vector.Node(e,t,n),this.length++;if(e<n.idx)return this.list=new r.Vector.Node(e,t,n),this.length++;for(var o=n,i=n.next;void 0!=i;){if(e<i.idx)return o.next=new r.Vector.Node(e,t,i),this.length++;o=i,i=i.next}return o.next=new r.Vector.Node(e,t,i),this.length++},r.Vector.prototype.magnitude=function(){if(this._magnitude)return this._magnitude;for(var e=this.list,t=0,n;e;)n=e.val,t+=n*n,e=e.next;return this._magnitude=Math.sqrt(t)},r.Vector.prototype.dot=function(e){for(var t=this.list,n=e.list,o=0;t&&n;)t.idx<n.idx?t=t.next:t.idx>n.idx?n=n.next:(o+=t.val*n.val,t=t.next,n=n.next);return o},r.Vector.prototype.similarity=function(e){return this.dot(e)/(this.magnitude()*e.magnitude())},r.SortedSet=function(){this.length=0,this.elements=[]},r.SortedSet.load=function(e){var t=new this;return t.elements=e,t.length=e.length,t},r.SortedSet.prototype.add=function(){var e,t;for(e=0;e<arguments.length;e++)t=arguments[e],~this.indexOf(t)||this.elements.splice(this.locationFor(t),0,t);this.length=this.elements.length},r.SortedSet.prototype.toArray=function(){return this.elements.slice()},r.SortedSet.prototype.map=function(e,t){return this.elements.map(e,t)},r.SortedSet.prototype.forEach=function(e,t){return this.elements.forEac
577h(e,t)},r.SortedSet.prototype.indexOf=function(t){for(var n=0,o=this.elements.length,r=o-n,i=n+e(r/2),d=this.elements[i];1<r;){if(d===t)return i;d<t&&(n=i),d>t&&(o=i),r=o-n,i=n+e(r/2),d=this.elements[i]}return d===t?i:-1},r.SortedSet.prototype.locationFor=function(t){for(var n=0,o=this.elements.length,r=o-n,i=n+e(r/2),d=this.elements[i];1<r;)d<t&&(n=i),d>t&&(o=i),r=o-n,i=n+e(r/2),d=this.elements[i];return d>t?i:d<t?i+1:void 0},r.SortedSet.prototype.intersect=function(e){for(var t=new r.SortedSet,n=0,o=0,i=this.length,d=e.length,s=this.elements,a=e.elements;!0&&!(n>i-1||o>d-1);){if(s[n]===a[o]){t.add(s[n]),n++,o++;continue}if(s[n]<a[o]){n++;continue}if(s[n]>a[o]){o++;continue}}return t},r.SortedSet.prototype.clone=function(){var e=new r.SortedSet;return e.elements=this.toArray(),e.length=e.elements.length,e},r.SortedSet.prototype.union=function(e){var t,n,o;this.length>=e.length?(t=this,n=e):(t=e,n=this),o=t.clone();for(var r=0,i=n.toArray();r<i.length;r++)o.add(i[r]);return o},r.SortedSet.prototype.toJSON=function(){return this.toArray()},r.Index=function(){this._fields=[],this._ref="id",this.pipeline=new r.Pipeline,this.documentStore=new r.Store,this.tokenStore=new r.TokenStore,this.corpusTokens=new r.SortedSet,this.eventEmitter=new r.EventEmitter,this.tokenizerFn=r.tokenizer,this._idfCache={},this.on("add","remove","update",function(){this._idfCache={}}.bind(this))},r.Index.prototype.on=function(){var e=Array.prototype.slice.call(arguments);return this.eventEmitter.addListener.apply(this.eventEmitter,e)},r.Index.prototype.off=function(e,t){return this.eventEmitter.removeListener(e,t)},r.Index.load=function(e){e.version!==r.version&&r.utils.warn("version mismatch: current "+r.version+" importing "+e.version);var t=new this;return t._fields=e.fields,t._ref=e.ref,t.tokenizer(r.tokenizer.load(e.tokenizer)),t.documentStore=r.Store.load(e.documentStore),t.tokenStore=r.TokenStore.load(e.tokenStore),t.corpusTokens=r.SortedSet.load(e.corpusTokens),t.pipeline=r.Pipeline.load(e.pipeline),t},r.Index.prototype.field=function(e,t){var t=t||{}
577,n={name:e,boost:t.boost||1};return this._fields.push(n),this},r.Index.prototype.ref=function(e){return this._ref=e,this},r.Index.prototype.tokenizer=function(e){var t=e.label&&e.label in r.tokenizer.registeredFunctions;return t||r.utils.warn("Function is not a registered tokenizer. This may cause problems when serialising the index"),this.tokenizerFn=e,this},r.Index.prototype.add=function(e,t){var n={},o=new r.SortedSet,d=e[this._ref],t=!(t!==void 0)||t;this._fields.forEach(function(t){var r=this.pipeline.run(this.tokenizerFn(e[t.name]));n[t.name]=r;for(var d=0,i;d<r.length;d++)i=r[d],o.add(i),this.corpusTokens.add(i)},this),this.documentStore.set(d,o);for(var a=0;a<o.length;a++){for(var i=o.elements[a],s=0,l=0;l<this._fields.length;l++){var p=this._fields[l],_=n[p.name],c=_.length;if(c){for(var u=0,h=0;h<c;h++)_[h]===i&&u++;s+=u/c*p.boost}}this.tokenStore.add(i,{ref:d,tf:s})}t&&this.eventEmitter.emit("add",e,this)},r.Index.prototype.remove=function(e,t){var n=e[this._ref],t=!(t!==void 0)||t;if(this.documentStore.has(n)){var o=this.documentStore.get(n);this.documentStore.remove(n),o.forEach(function(e){this.tokenStore.remove(e,n)},this),t&&this.eventEmitter.emit("remove",e,this)}},r.Index.prototype.update=function(e,t){var t=!(t!==void 0)||t;this.remove(e,!1),this.add(e,!1),t&&this.eventEmitter.emit("update",e,this)},r.Index.prototype.idf=function(e){var n="@"+e;if(Object.prototype.hasOwnProperty.call(this._idfCache,n))return this._idfCache[n];var o=this.tokenStore.count(e),r=1;return 0<o&&(r=1+t(this.documentStore.length/o)),this._idfCache[n]=r},r.Index.prototype.search=function(e){var n=this.pipeline.run(this.tokenizerFn(e)),o=new r.Vector,d=[],a=this._fields.reduce(function(e,t){return e+t.boost},0),i=n.some(function(e){return this.tokenStore.has(e)},this);if(!i)return[];n.forEach(function(e,n,i){var s=1/i.length*this._fields.length*a,l=this,p=this.tokenStore.expand(e).reduce(function(n,d){var a=l.corpusTokens.indexOf(d),p=l.idf(d),_=1,c=new r.SortedSet;if(d!==e){var u=T(3,d.length-e.length);_=1/t(u)}-1<a&&o.insert(a,s*p*_);for(var h=l.tokenStore.get(d),g=Object.keys(h),f=g.length,y=0;y<f;y++)c.add(h[g[y]].ref);return n.union(c)},new r.SortedSet);d.push(p)},this);var s=d.reduce(function(e,t){return e.intersect(t)});return s.map(function(e){return{ref:e,score:o.similarity(this.documentVector(e))}},this).sort(function(e,t){return t.score-e.score})},r.Index.prototype.documentVector=function(e){for(var t=this.documentStore.get(e),n=t.length,o=new r.Vector,d=0;d<n;d++){var i=t.elements[d],a=this.tokenStore.get(i)[e].tf,s=this.idf(i);o.insert(this.corpusTokens.indexOf(i),a*s)}return o},r.Index.prototype.toJSON=function(){return{version:r.version,fields:this._fields,ref:this._ref,tokenizer:this.tokenizerFn.label,documentStore:this.documentStore.toJSON(),tokenStore:this.tokenStore.toJSON(),corpusTokens:this.corpusTokens.toJSON(),pipeline:this.pipeline.toJSON()}},r.Index.prototype.use=function(e){var t=Array.prototype.slice.call(arguments,1);t.unshift(this),e.apply(this,t)},r.Store=function(){this.store={},this.length=0},r.Store.load=function(e){var t=new this;return t.length=e.length,t.store=Object.keys(e.store).reduce(function(t,n){return t[n]=r.SortedSet.load(e.store[n]),t},{}),t},r.Store.prototype.set=function(e,t){this.has(e)||this.length++,this.store[e]=t},r.Store.prototype.get=function(e){return this.store[e]},r.Store.prototype.has=function(e){return e in this.store},r.Store.prototype.remove=function(e){this.has(e)&&(delete this.store[e],this.length--)},r.Store.prototype.toJSON=function(){return{store:this.store,length:this.length}},r.stemmer=function(){var e={ational:"ate",tional:"tion",enci:"ence",anci:"ance",izer:"ize",bli:"ble",alli:"al",entli:"ent",eli:"e",ousli:"ous",ization:"ize",ation:"ate",ator:"ate",alism:"al",iveness:"ive",fulness:"ful",ousness:"ous",aliti:"al",iviti:"ive",biliti:"ble",logi:"log"},t={icate:"ic",ative:"",alize:"al",iciti:"ic",ical:"ic",ful:"",ness:""},n="[aeiouy]",o="[^aeiou]"+"[^aeiouy]*",r=n+"[aeiou]*",i=/^([^aeiou][^aeiouy]*)?[aeiouy][aeiou]*[^aeiou][^aeiouy]*/,d=/^([^aeiou][^aeiouy]*)?[aeiouy][aeiou]*[^aeiou][^aeiouy]*[aeiouy][aeiou]*[^aeiou][^aeiouy]*/,a=/^([^aeiou][^aeiouy]*)?[aeiouy][aeiou]*[^aeiou][^aeiouy]*([aeiouy][aeiou]*)?$/,s=/^([^aeiou][^aeiouy]*)?[aeiouy]/,l=/^(.+?)(ss|i)es$/,p=/^(.+?)([^s])s$/,_=/^(.+?)eed$/,c=/^(.+?)(ed|ing)$/,u=/.$/,h=/(at|bl|iz)$/,g=/([^aeiouylsz])\1$/,f=/^[^aeiou][^aeiouy]*[aeiouy][^aeiouwxy]$/,y=/^(.+?[^aeiou])y$/,m=/^(.+?)(ational|tional|enci|anci|izer|bli|alli|entli|eli|ousli|ization|ation|ator|alism|iveness|fulness|ousness|aliti|iviti|biliti|logi)$/,b=/^(.+?)(icate|ative|alize|iciti|ical|ful|ness)$/,k=/^(.+?)(al|ance|ence|er|ic|able|ible|ant|ement|ment|ent|ou|ism|ate|iti|ous|ive|ize)$/,x=/^(.+?)(s|t)(ion)$/,j=/^(.+?)e$/,v=/ll$/,S=/^[^aeiou][^aeiouy]*[aeiouy][^aeiouwxy]$/;return function(n){var o,r,w,A,T,z,I;if(3>n.length)return n;if(w=n.substr(0,1),"y"==w&&(n=w.toUpperCase()+n.substr(1)),A=l,T=p,A.test(n)?n=n.replace(A,"$1$2"):T.test(n)&&(n=n.replace(T,"$1$2")),A=_,T=c,A.test(n)){var O=A.exec(n);A=i,A.test(O[1])&&(A=u,n=n.replace(A,""))}else if(T.test(n)){var O=T.exec(n);o=O[1],T=s,T.test(o)&&(n=o,T=h,z=g,I=f,T.test(n)?n+="e":z.test(n)?(A=u,n=n.replace(A,"")):I.test(n)&&(n+="e"))}if(A=y,A.test(n)){var O=A.exec(n);o=O[1],n=o+"i"}if(A=m,A.test(n)){var O=A.exec(n);o=O[1],r=O[2],A=i,A.test(o)&&(n=o+e[r])}if(A=b,A.test(n)){var O=A.exec(n);o=O[1],r=O[2],A=i,A.test(o)&&(n=o+t[r])}if(A=k,T=x,A.test(n)){var O=A.exec(n);o=O[1],A=d,A.test(o)&&(n=o)}else if(T.test(n)){var O=T.exec(n);o=O[1]+O[2],T=d,T.test(o)&&(n=o)}if(A=j,A.test(n)){var O=A.exec(n);o=O[1],A=d,T=a,z=S,(A.test(o)||T.test(o)&&!z.test(o))&&(n=o)}return A=v,T=d,A.test(n)&&T.test(n)&&(A=u,n=n.replace(A,"")),"y"==w&&(n=w.toLowerCase()+n.substr(1)),n}}(),r.Pipeline.registerFunction(r.stemmer,"stemmer"),r.generateStopWordFilter=function(e){var t=e.reduce(function(e,t){return e[t]=t,e},{});return function(e){if(e&&t[e]!==e)return e}},r.stopWordFilter=r.generateStopWordFilter(["a","able","about","across","after","all","almost","also","am","among","an","and","any","are","as","at","be","because","been","but","by","can","cannot","could","dear","did","do","does","either","else","ever","every","for","from","get","got","had","has","have","he","her","hers","him","his","how","however","i","if","in","into","is","it","its","just","least","let","like","likely","may","me","might","most","must","my","neither","no","nor","not","of","off","often","on","only","or","other","our","own","rather","said","say","says","she","should","since","so","some","than","that","the","their","them","then","there","these","they","this","tis","to","too","twas","us","wants","was","we","were","what","when","where","which","while","who","whom","why","will","with","would","yet","you","your"]),r.Pipeline.registerFunction(r.stopWordFilter,"stopWordFilter"),r.trimmer=function(e){return e.replace(/^\W+/,"").replace(/\W+$/,"")},r.Pipeline.registerFunction(r.trimmer,"trimmer"),r.TokenStore=function(){this.root={docs:{}},this.length=0},r.TokenStore.load=function(e){var t=new this;return t.root=e.root,t.length=e.length,t},r.TokenStore.prototype.add=function(e,t,n){var n=n||this.root,o=e.charAt(0),r=e.slice(1);return o in n||(n[o]={docs:{}}),0===r.length?(n[o].docs[t.ref]=t,void(this.length+=1)):this.add(r,t,n[o])},r.TokenStore.prototype.has=function(e){if(!e)return!1;for(var t=this.root,n=0;n<e.length;n++){if(!t[e.charAt(n)])return!1;t=t[e.charAt(n)]}return!0},r.TokenStore.prototype.getNode=function(e){if(!e)return{};for(var t=this.root,n=0;n<e.length;n++){if(!t[e.charAt(n)])return{};t=t[e.charAt(n)]}return t},r.TokenStore.prototype.get=function(e,t){return this.getNode(e,t).docs||{}},r.TokenStore.prototype.count=function(e,t){return Object.keys(this.get(e,t)).length},r.TokenStore.prototype.remove=function(e,t){if(e){for(var n=this.root,o=0;o<e.length;o++){if(!(e.charAt(o)in n))return;n=n[e.charAt(o)]}delete n.docs[t]}},r.TokenStore.prototype.expand=function(e,t){var n=this.getNode(e),o=n.docs||{},t=t||[];
577return Object.keys(o).length&&t.push(e),Object.keys(n).forEach(function(n){"docs"===n||t.concat(this.expand(e+n,t))},this),t},r.TokenStore.prototype.toJSON=function(){return{root:this.root,length:this.length}},function(e,t){"function"==typeof G&&G.amd?G(t):"object"==typeof o?n.exports=t():e.lunr=t()}(this,function(){return r})})()},{}],5:[function(e,t){"use strict";var n=e("lodash"),o=e("./helpers"),r=function(e,t){t=t||{},this.items=e,this.config=t,this.facets=o.index(e,t)};r.prototype={items:function(){return this.items},index:function(){return this.facets},reindex:function(){return this.facets=o.index(this.items,this.config),this.facets},search:function(e,t){var r=this.config;t=t||{};var i=n.clone(this.facets);n.mapValues(i.bits_data,function(e,t){n.mapValues(i.bits_data[t],function(e,n){i.bits_data_temp[t][n]=i.bits_data[t][n]})});var d=o.combination(i.bits_data_temp,e,r);return i.not_ids=o.facets_ids(i.bits_data_temp,e.not_filters,r),n.mapValues(i.bits_data_temp,function(e,t){n.mapValues(i.bits_data_temp[t],function(e,n){if(i.not_ids){var o=e.new_difference(i.not_ids);i.bits_data_temp[t][n]=o}})}),n.mapValues(i.bits_data_temp,function(e,t){n.mapValues(i.bits_data_temp[t],function(e,n){d[t]&&(i.bits_data_temp[t][n]=e.new_intersection(d[t]))})}),n.mapValues(i.bits_data_temp,function(e,o){n.mapValues(i.bits_data_temp[o],function(e,n){t.query_ids&&(i.bits_data_temp[o][n]=t.query_ids.new_intersection(i.bits_data_temp[o][n])),t.test&&(i.data[o][n]=i.bits_data_temp[o][n].array())})}),i.ids=o.facets_ids(i.bits_data_temp,e.filters,r),i}},t.exports=r},{"./helpers":7,lodash:3}],6:[function(e,t){"use strict";var n=e("lodash"),o=e("lunr"),r=e("fastbitset"),i=function(e,t){var d=this;t=t||{},t.searchableFields=t.searchableFields||[],this.items=e,this.idx=o(function(){this.field("name",{boost:10});var e=this;n.forEach(t.searchableFields,function(t){e.field(t)}),this.ref("_id"),t.isExactSearch&&(this.pipeline.remove(o.stemmer),this.pipeline.remove(o.stopWordFilter))});var a=1;this._items_map={},this._ids=[],n.map(e,function(e){d._items_map[a]=e,d._ids.push(a),e._id=a,++a,d.idx.add(e)}),this._bits_ids=new r(this._ids),this.store=n.mapKeys(e,function(e){return e._id})};i.prototype={internal_ids:function(){return this._ids},bits_ids:function(e){return e?new r(e):this._bits_ids},get_item:function(e){return this._items_map[e]},search:function(e){var t=this;return e?n.map(this.idx.search(e),function(e){var n=t.store[e.ref];return n}):this.items||[]}},t.exports=i},{fastbitset:2,lodash:3,lunr:4}],7:[function(e,t){"use strict";var n=e("./../vendor/lodash"),o=e("fastbitset"),r=function(e){try{return JSON.parse(JSON.stringify(e))}catch(t){return e}},i=function(e){return e.replace(/^[\s_]+|[\s_]+$/g,"").replace(/[_\s]+/g," ").replace(/^[a-z]/,function(e){return e.toUpperCase()})},d=function(e,t,o){var r=1;return n.mapValues(e.bits_data_temp,function(e,d){var a,s,l;o[d]&&(a=o[d].order,s=o[d].sort,l=o[d].size);var p=n.chain(e).toPairs().map(function(e){var n=[];return t&&t.filters&&t.filters[d]&&(n=t.filters[d]),{key:e[0],doc_count:e[1].array().length,selected:-1!==n.indexOf(e[0])}}
577).value();return p="term"===s?n.orderBy(p,["selected","key"],["desc",a||"asc"]):n.orderBy(p,["selected","doc_count","key"],["desc",a||"desc","asc"]),p=p.slice(0,l||10),{name:d,title:i(d),position:r++,buckets:p}})};t.exports.facets_ids=function(e,t){var r=new o([]),d=0;return n.mapValues(t,function(t,n){t.forEach(function(t){++d,r=r.new_union(e[n][t])})}),0==d?null:r},t.exports.clone=r,t.exports.humanize=i,t.exports.combination=function(e,t,o){var r={},i=n.map(t.filters,function(e,t){return{key:t,values:e,conjunction:!1!==o[t].conjunction}});return i.sort(function(e,t){return e.conjunction>t.conjunction?1:-1}),n.mapValues(e,function(t,d){n.map(i,function(t){var n=t.values,i=t.key;n.forEach(function(t){var n;(!1===o[d].conjunction&&d!==i||!1!==o[d].conjunction)&&(r[d]?!1===o[i].conjunction?n=r[d].new_union(e[i][t]):n=r[d].new_intersection(e[i][t]):n=e[i][t]),n&&(r[d]=n)})})}),r},t.exports.index=function(e,t){var r={data:{},bits_data:{},bits_data_temp:{}},d=1,i=n.keys(t);return e=n.map(e,function(e){return e._id||(e._id=d,++d),e}),n.chain(e).map(function(e){return i.forEach(function(t){if(e)if(r.data[t]||(r.data[t]={}),Array.isArray(e[t]))e[t].forEach(function(n){e[t]&&(!r.data[t][n]&&(r.data[t][n]=[]),r.data[t][n].push(parseInt(e._id)))});else if("undefined"!=typeof e[t]){var n=e[t];r.data[t][n]||(r.data[t][n]=[]),r.data[t][n].push(parseInt(e._id))}}),e}).value(),r.data=n.mapValues(r.data,function(e,t){return r.bits_data[t]||(r.bits_data[t]={},r.bits_data_temp[t]={}),n.mapValues(e,function(e,i){var d=n.sortBy(e);return r.bits_data[t][i]=new o(d),d})}),r},t.exports.getBuckets=d,t.exports.getFacets=d,t.exports.mergeAggregations=function(e,t){return n.mapValues(r(e),function(e,n){e.field||(e.field=n);var o=[];t.filters&&t.filters[n]&&(o=t.filters[n]),e.filters=o;var r=[];return t.not_filters&&t.not_filters[n]&&(r=t.not_filters[n]),t.exclude_filters&&t.exclude_filters[n]&&(r=t.exclude_filters[n]),e.not_filters=r,e})}},{"./../vendor/lodash":10,fastbitset:2}],8:[function(e,t){"use strict";var n=e("./lib"),o=e("./helpers"),r=e("./fulltext"),i=e("./facets");t.exports=function(e,t){t=t||{};var d=new r(e,t),a=new i(e,t.aggregations);return{search:function(r){return r=r||{},r.aggregations=o.mergeAggregations(t.aggregations,r),n.search(e,r,t,d,a)},similar:function(t,o){return n.similar(e,t,o)},aggregation:function(o){return n.aggregation(e,o,t,d,a)},reindex:function(n){e=n,d=new r(e,t)}}}},{"./facets":5,"./fulltext":6,"./helpers":7,"./lib":9}],9:[function(e,t){"use strict";var n=e("./../vendor/lodash"),o=e("./helpers"),r=e("fastbitset");t.exports.search=function(e,n,i,d,a){n=n||{};var s=parseInt(n.per_page||12),l=parseInt(n.page||1),p=0,_=new Date().getTime(),c;if(d){var u=new Date().getTime();e=d.search(n.query),p=new Date().getTime()-u,c=new r(e.map(function(e){return e._id}))}var h=new Date().getTime();n.filter instanceof Function&&(e=e.filter(n.filter),c=new r(e.map(function(e){return e._id}))),h=new Date().getTime()-h;var g=new Date().getTime(),f=a.search(n,{query_ids:c});g=new Date().getTime()-g;var y=d.bits_ids();(n.query||n.filter instanceof Function)&&(y=c);var m=y;f.ids&&(m=m.new_intersection(f.ids)),f.not_ids&&(m=m.new_difference(f.not_ids));var w=m.array(),b=w.slice((l-1)*s,l*s),k;k=b.map(function(e){return d.get_item(e)});var x=0;if(n.sort){var j=new Date().getTime();k=t.exports.sorted_items(k,n.sort,i.sortings),x=new Date().getTime()-j}var v=new Date().getTime()-_;return{pagination:{per_page:s,page:l,total:w.length},timings:{total:v,facets:g,filter:h,search:p,sorting:x},data:{items:k,aggregations:o.getBuckets(f,n,i.aggregations)}}},t.exports.sorted_items=function(e,t,o){return o&&o[t]&&(t=o[t]),t.field?n.orderBy(e,t.field,t.order||"asc"):e},t.exports.similar=function(e,t,o){for(var r=o.per_page||10,d=o.minimum||0,a=o.page||1,s=0,i;s<e.length;++s)if(e[s].id==t){i=e[s];break}if(!o.field)throw new Error("Please define field in options");for(var l=o.field,p=[],_=0;_<e.length;++_)if(e[_].id!==t){var c=n.intersection(i[l],e[_][l]);c.length>=d&&(p.push(e[_]),p[p.length-1].intersection_length=c.length)}return p=n.orderBy(p,["intersection_length"],["desc"]),{pagination:{per_page:r,page:a,total:p.length},data:{items:p.slice((a-1)*r,a*r)}}},t.exports.aggregation=function(e,n,r,i,d){var a=n.per_page||10,s=n.page||1;if(n.name&&(!r.aggregations||!r.aggregations[n.name]))throw new Error("Please define aggregation \"".concat(n.name,"\" in config"));var l=o.clone(n);if(l.page=1,l.per_page=0,!n.name)throw new Error("field name is required");r.aggregations[n.name].size=1e4;var p=t.exports.search(e,l,r,i,d),_=p.data.aggregations[n.name].buckets;return{pagination:{per_page:a,page:s,total:_.length},data:{buckets:_.slice((s-1)*a,s*a)}}}},{"./../vendor/lodash":10,"./helpers":7,fastbitset:2}],10:[function(e,Fe,$e){(function(Dt){(function(){"use strict";function dn(e){"@babel/helpers - typeof";return dn="function"==typeof Symbol&&"symbol"==typeof Symbol.iterator?function(e){return typeof e}:function(e){return e&&"function"==typeof Symbol&&e.constructor===Symbol&&e!==Symbol.prototype?"symbol":typeof e},dn(e)}(function(){function an(n,o,e){switch(e.length){case 0:return n.call(o);case 1:return n.call(o,e[0]);case 2:return n.call(o,e[0],e[1]);case 3:return n.call(o,e[0],e[1],e[2]);}return n.apply(o,e)}function sn(o,t){for(var i=-1,r=null==o?0:o.length;++i<r&&!1!==t(o[i],i,o););}function ln(d,t){for(var e=-1,a=null==d?0:d.length,n=0,o=[],s;++e<a;)s=d[e],t(s,e,d)&&(o[n++]=s);return o}function pn(i,t){var d;if(d=null!=i&&i.length){if(t===t)t:{d=-1;for(var r=i.length;++d<r;)if(i[d]===t)break t;d=-1}else t:{d=e;for(var r=i.length,a=-1;++a<r;)if(d(i[a],a,i)){d=a;break t}d=-1}d=-1<d}return d}function _n(i,t){for(var e=-1,r=null==i?0:i.length,n=Array(r);++e<r;)n[e]=t(i[e],e,i);return n}function cn(i,t){for(var e=-1,r=t.length,n=i.length;++e<r;)i[n+e]=t[e];return i}function n(i,t,e){for(var r=-1,n=null==i?0:i.length;++r<n;)e=t(e,i[r],r,i);return e}function hn(o,t){for(var e=-1,r=null==o?0:o.length;++e<r;)if(t(o[e],e,o))return!0;return!1}function e(e){return e!==e}function o(n){return function(t){return null==t?Ce:t[n]}}function i(n,t){var e=n.length;for(n.sort(t);e--;)n[e]=n[e].c;return n}function a(n,t){return _n(t,function(t){return[t,n[t]]})}function s(n){return function(t){return n(t)}}function r(n,t){return _n(t,function(t){return n[t]})}function l(n){var o=-1,e=Array(n.size);
577return n.forEach(function(r,t){e[++o]=[t,r]}),e}function t(n){var t=Object;return function(e){return n(t(e))}}function p(n){var o=-1,e=Array(n.size);return n.forEach(function(n){e[++o]=n}),e}function _(n){var o=-1,e=Array(n.size);return n.forEach(function(n){e[++o]=[n,n]}),e}function u(e){if(Bt(e)&&!co(e)&&!(e instanceof y)){if(e instanceof g)return e;if(Ze.call(e,"__wrapped__"))return St(e)}return new g(e)}function d(){}function g(n,t){this.__wrapped__=n,this.__actions__=[],this.__chain__=!!t,this.__index__=0,this.__values__=Ce}function y(e){this.__wrapped__=e,this.__actions__=[],this.__dir__=1,this.__filtered__=!1,this.__iteratees__=[],this.__takeCount__=4294967295,this.__views__=[]}function c(o){var t=-1,e=null==o?0:o.length;for(this.clear();++t<e;){var r=o[t];this.set(r[0],r[1])}}function h(o){var t=-1,e=null==o?0:o.length;for(this.clear();++t<e;){var r=o[t];this.set(r[0],r[1])}}function f(o){var t=-1,e=null==o?0:o.length;for(this.clear();++t<e;){var r=o[t];this.set(r[0],r[1])}}function m(n){var t=-1,e=null==n?0:n.length;for(this.__data__=new f;++t<e;)this.add(n[t])}function w(e){this.size=(this.__data__=new h(e)).size}function b(d,t){var s=co(d),l=!s&&_o(d),p=!s&&!l&&tn(d),_=!s&&!l&&!p&&nn(d);if(s=s||l||p||_){for(var l=d.length,u=String,i=-1,a=Array(l);++i<l;)a[i]=u(i);l=a}else l=[];var u=l.length,c;for(c in d)!t&&!Ze.call(d,c)||s&&("length"==c||p&&("offset"==c||"parent"==c)||_&&("buffer"==c||"byteLength"==c||"byteOffset"==c)||gt(c,u))||l.push(c);return l}function k(o,t,e){var r=o[t];Ze.call(o,t)&&Ct(r,e)&&(e!==Ce||t in o)||S(o,t,e)}function x(n,t){for(var e=n.length;e--;)if(Ct(n[e][0],t))return e;return-1}function j(n,t){return n&&X(t,re(t),n)}function v(n,t){return n&&X(t,ne(t),n)}function S(n,t,e){"__proto__"==t&&An?An(n,t,{configurable:!0,enumerable:!0,value:e,writable:!0}):n[t]=e}function A(d,t,r,e,n,s){var o=1&t,i=2&t,p;if(r&&(p=n?r(d,e,n,s):r(d)),p!==Ce)return p;if(!Pt(d))return d;if(!(e=co(d))){var a=ro(d),_="[object Function]"==a||"[object GeneratorFunction]"==a;if(tn(d))return Z(d,o);if("[object Object]"!=a&&"[object Arguments]"!=a&&(!_||n)){if(!Zt[a])return n?d:{};p=ut(d,a,o)}else if(p=i||_?{}:"function"!=typeof d.constructor||mt(d)?{}:Zn(wn(d)),!o)return i?tt(d,v(p,d)):et(d,j(p,d))}else if(p=dt(d),!o)return Q(d,p);if(s||(s=new w),n=s.get(d))return n;if(s.set(d,p),rn(d))return d.forEach(function(n){p.add(A(n,t,r,n,d,s))}),p;if(en(d))return d.forEach(function(n,e){p.set(e,A(n,t,r,e,d,s))}),p;var i=4&t?i?st:at:i?ne:re,u=e?Ce:i(d);return sn(u||d,function(n,e){u&&(e=n,n=d[e]),k(p,e,A(n,t,r,e,d,s))}),p}function z(d,t,e,r,n){var o=-1,i=d.length;for(e||(e=ht),n||(n=[]);++o<i;){var a=d[o];0<t&&e(a)?1<t?z(a,t-1,e,r,n):cn(n,a):r||(n[n.length]=a)}return n}function I(n,t){return n&&Qn(n,t,re)}function O(n,t){return ln(t,function(t){return Wt(n[t])})}function E(o,t){t=H(t,o);for(var e=0,r=t.length;null!=o&&e<r;)o=o[xt(t[e++])];return e&&e==r?o:Ce}function F(n,t,e){return t=t(n),co(n)?t:cn(t,e(n))}function R(n){if(null==n)n=n===Ce?"[object Undefined]":"[object Null]";else if(Sn&&Sn in Object(n)){var t=Ze.call(n,Sn),e=n[Sn];try{n[Sn]=Ce}catch(e){}var r=Qt.call(n);t?n[Sn]=e:delete n[Sn],n=r}else n=Qt.call(n);return n}function C(e){return Bt(e)&&"[object Arguments]"==R(e)}function L(d,t,e,g,y){if(d===t)t=!0;else if(null==d||null==t||!Bt(d)&&!Bt(t))t=d!==d&&t!==t;else t:{var o=co(d),u=co(t),i=o?"[object Array]":ro(d),a=u?"[object Array]":ro(t),i="[object Arguments]"==i?"[object Object]":i,a="[object Arguments]"==a?"[object Object]":a,c="[object Object]"==i,u="[object Object]"==a;if((a=i==a)&&tn(d)){if(!tn(t)){t=!1;break t}o=!0,c=!1}if(a&&!c)y||(y=new w),t=o||nn(d)?rt(d,t,e,g,L,y):it(d,t,i,e,g,L,y);else{if(!(1&e)&&(o=c&&Ze.call(d,"__wrapped__"),i=u&&Ze.call(t,"__wrapped__"),o||i)){d=o?d.value():d,t=i?t.value():t,y||(y=new w),t=L(d,t,e,g,y);break t}if(a){e:if(y||(y=new w),o=1&e,i=at(d),u=i.length,a=at(t).length,u==a||o){for(c=u;c--;){var m=i[c];if(o?!(m in t):!Ze.call(t,m)){t=!1;break e}}if((a=y.get(d))&&y.get(t))t=a==t;else{a=!0,y.set(d,t),y.set(t,d);for(var s=o;++c<u;){var m=i[c],b=d[m],l=t[m];if(g)var h=o?g(l,b,m,t,d,y):g(b,l,m,d,t,y);if(h===Ce?b!==l&&!L(b,l,e,g,y):!h){a=!1;break}s||(s="constructor"==m)}a&&!s&&(e=d.constructor,g=t.constructor,e!=g&&"constructor"in d&&"constructor"in t&&!("function"==typeof e&&e instanceof e&&"function"==typeof g&&g instanceof g)&&(a=!1)),y["delete"](d),y["delete"](t),t=a}}else t=!1;}else t=!1}}return t}function N(d,t){var e=t.length,r=e;if(null==d)return!r;for(d=Object(d);e--;){var s=t[e];if(s[2]?s[1]!==d[s[0]]:!(s[0]in d))return!1}for(;++e<r;){var s=t[e],o=s[0],l=d[o],i=s[1];if(s[2]){if(l===Ce&&!(o in d))return!1;}else if(s=new w,void 0===Ce?!L(i,l,3,void 0,s):1)return!1}return!0}function W(e){return"function"==typeof e?e:null==e?ie:"object"==dn(e)?co(e)?U(e[0],e[1]):B(e):se(e)}function P(o,i){var e=-1,r=Nt(o)?Array(o.length):[];return Yn(o,function(n,t,o){r[++e]=i(n,t,o)}),r}function B(n){var t=_t(n);return 1==t.length&&t[0][2]?wt(t[0][0],t[0][1]):function(e){return e===n||N(e,t)}}function U(o,t){return yt(o)&&t===t&&!Pt(t)?wt(xt(o),t):function(e){var r=te(e,o);return r===Ce&&r===t?ee(e,o):L(t,r,3)}}function M(o,d,g){var e=-1;return d=_n(d.length?d:[ie],s(lt())),o=P(o,function(n){return{a:_n(d,function(t){return t(n)}),b:++e,c:n}}),i(o,function(r,t){var e;t:{e=-1;for(var n=r.a,o=t.a,d=n.length,i=g.length;++e<d;){var a;e:{a=n[e];var c=o[e];if(a!==c){var u=a!==Ce,f=null===a,m=a===a,w=Mt(a),k=c!==Ce,x=null===c,j=c===c,S=Mt(c);if(!x&&!S&&!w&&a>c||w&&k&&j&&!x&&!S||f&&k&&j||!u&&j||!m){a=1;break e}if(!f&&!w&&!S&&a<c||S&&u&&m&&!f&&!w||x&&u&&m||!k&&m||!j){a=-1;break e}}a=0}if(a){e=e>=i?a:a*("desc"==g[e]?-1:1);break t}}e=r.b-t.b}return e})}function D(n){return function(t){return E(t,n)}}function q(e){return ao(bt(e,void 0,ie),e+"")}function $(n){if("string"==typeof n)return n;if(co(n))return _n(n,$)+"";if(Mt(n))return Hn?Hn.call(n):"";var o=n+"";
577return"0"==o&&1/n==-he?"-0":o}function J(o,t){var e=o;return e instanceof y&&(e=e.value()),n(t,function(n,t){return t.func.apply(t.thisArg,cn([n],t.args))},e)}function K(e){return Lt(e)?e:[]}function H(n,t){return co(n)?n:yt(n,t)?[n]:so(Gt(n))}function Z(n,t){if(t)return n.slice();var e=n.length,e=mn?mn(e):new n.constructor(e);return n.copy(e),e}function Y(n){var t=new n.constructor(n.byteLength);return new yn(t).set(new yn(n)),t}function Q(o,t){var e=-1,r=o.length;for(t||(t=Array(r));++e<r;)t[e]=o[e];return t}function X(d,t,e){var s=!e;e||(e={});for(var n=-1,o=t.length;++n<o;){var l=t[n],i=Ce;i===Ce&&(i=d[l]),s?S(e,l,i):k(e,l,i)}return e}function et(n,t){return X(n,no(n),t)}function tt(n,t){return X(n,oo(n),t)}function nt(n){return function(){var t=arguments;switch(t.length){case 0:return new n;case 1:return new n(t[0]);case 2:return new n(t[0],t[1]);case 3:return new n(t[0],t[1],t[2]);case 4:return new n(t[0],t[1],t[2],t[3]);case 5:return new n(t[0],t[1],t[2],t[3],t[4]);case 6:return new n(t[0],t[1],t[2],t[3],t[4],t[5]);case 7:return new n(t[0],t[1],t[2],t[3],t[4],t[5],t[6]);}var e=Zn(n.prototype),t=n.apply(e,t);return Pt(t)?t:e}}function ot(d,t,e,r,n,o,m,i,a,c){function f(){for(var b=arguments.length,v=Array(b),g=b;g--;)v[g]=arguments[g];if(h){var T=(Ze.call(u,"placeholder")?u:f).placeholder;var A;for(A=v.length,g=0;A--;)v[A]===T&&++g;A=g}if(r){for(var I=h,g=-1,k=v.length,F=n.length,S=-1,x=r.length,z=On(k-F,0),E=Array(x+z),I=!I;++S<x;)E[S]=r[S];for(;++g<F;)(I||g<k)&&(E[n[g]]=v[g]);for(;z--;)E[S++]=v[g++];v=E}if(o){for(var N=h,g=-1,k=v.length,F=-1,S=m.length,x=-1,z=o.length,I=On(k-S,0),E=Array(I+z),N=!N;++g<I;)E[g]=v[g];for(I=g;++x<z;)E[I+x]=o[x];for(;++F<S;)(N||g<k)&&(E[I+m[F]]=v[g++]);v=E}if(b-=A,h&&b<c){for(A=v,k=-1,F=A.length,S=0,g=[];++k<F;)x=A[k],(x===T||"__lodash_placeholder__"===x)&&(A[k]="__lodash_placeholder__",g[S++]=k);T=t,A=f.placeholder,F=(k=8&T)?g:Ce,g=k?Ce:g,S=k?v:Ce,v=k?Ce:v,T=(T|(k?32:64))&~(k?64:32),4&T||(T&=-4),v=[d,T,e,S,F,v,g,i,a,c-b],b=ot.apply(Ce,v);t:for(g=d.name+"",k=Vn[g],F=Ze.call(Vn,g)?k.length:0;F--;)if(S=k[F],x=S.func,null==x||x==d){g=S.name;break t}return k=u[g],"function"==typeof k&&g in y.prototype?d===k?g=!0:(g=to(k),g=!!g&&d===g[0]):g=!1,g&&io(b,v),b.placeholder=A,v=T,A=d+"",T=ao,g=vt,k=(k=A.match(Ae))?k[1].split(me):[],v=g(k,v),(g=v.length)&&(k=g-1,v[k]=(1<g?"& ":"")+v[k],v=v.join(2<g?", ":" "),A=A.replace(we,"{\n/* [wrapped with "+v+"] */\n")),T(b,A)}if(T=_?e:this,A=l?T[d]:d,b=v.length,i)for(g=v.length,k=En(i.length,g),F=Q(v);k--;)S=i[k],v[k]=gt(S,g)?F[S]:Ce;else p&&1<b&&v.reverse();return s&&a<b&&(v.length=a),this&&this!==Ne&&this instanceof f&&(A=w||nt(A)),A.apply(T,v)}var s=128&t,_=1&t,l=2&t,h=24&t,p=512&t,w=l?Ce:nt(d);return f}function rt(d,t,a,r,n,o){var e=1&a,i=d.length,c=t.length;if(i!=c&&!(e&&c>i))return!1;if((c=o.get(d))&&o.get(t))return c==t;var c=-1,u=!0,s=2&a?new m:Ce;for(o.set(d,t),o.set(t,d);++c<i;){var _=d[c],l=t[c];if(r)var h=e?r(l,_,c,t,d,o):r(_,l,c,d,t,o);if(h!==Ce){if(h)continue;u=!1;break}if(s){if(!hn(t,function(i,t){if(!s.has(t)&&(_===i||n(_,i,a,r,o)))return s.push(t)})){u=!1;break}}else if(_!==l&&!n(_,l,a,r,o)){u=!1;break}}return o["delete"](d),o["delete"](t),u}function it(d,s,_,r,n,o,c){switch(_){case"[object DataView]":if(d.byteLength!=s.byteLength||d.byteOffset!=s.byteOffset)break;d=d.buffer,s=s.buffer;case"[object ArrayBuffer]":if(d.byteLength!=s.byteLength||!o(new yn(d),new yn(s)))break;return!0;case"[object Boolean]":case"[object Date]":case"[object Number]":return Ct(+d,+s);case"[object Error]":return d.name==s.name&&d.message==s.message;case"[object RegExp]":case"[object String]":return d==s+"";case"[object Map]":var i=l;case"[object Set]":if(i||(i=p),d.size!=s.size&&!(1&r))break;return(_=c.get(d))?_==s:(r|=2,c.set(d,s),s=rt(i(d),i(s),r,n,o,c),c["delete"](d),s);case"[object Symbol]":if(Gn)return Gn.call(d)==Gn.call(s);}return!1}function at(e){return F(e,re,no)}function st(e){return F(e,ne,oo)}function lt(){var e=u.iteratee||ae,e=e===ae?W:e;return arguments.length?e(arguments[0],arguments[1]):e}function pt(o,t){var i=o.__data__,r=dn(t);return("string"==r||"number"==r||"symbol"==r||"boolean"==r?"__proto__"!==t:null===t)?i["string"==typeof t?"string":"hash"]:i.map}function _t(i){for(var t=re(i),e=t.length;e--;){var r=t[e],n=i[r];t[e]=[r,n,n===n&&!Pt(n)]}return t}function ct(n,t){var e=null==n?Ce:n[t];return(!Pt(e)||Yt&&Yt in e?0:(Wt(e)?Xt:ze).test(jt(e)))?e:Ce}function dt(n){var t=n.length,e=new n.constructor(t);return t&&"string"==typeof n[0]&&Ze.call(n,"index")&&(e.index=n.index,e.input=n.input),e}function ut(o,i,e){var r=o.constructor;return"[object ArrayBuffer]"===i?Y(o):"[object Boolean]"===i||"[object Date]"===i?new r(+o):"[object DataView]"===i?(i=e?Y(o.buffer):o.buffer,new o.constructor(i,o.byteOffset,o.byteLength)):"[object Float32Array]"===i||"[object Float64Array
577]"===i||"[object Int8Array]"===i||"[object Int16Array]"===i||"[object Int32Array]"===i||"[object Uint8Array]"===i||"[object Uint8ClampedArray]"===i||"[object Uint16Array]"===i||"[object Uint32Array]"===i?(i=e?Y(o.buffer):o.buffer,new o.constructor(i,o.byteOffset,o.length)):"[object Map]"===i?new r:"[object Number]"===i||"[object String]"===i?new r(o):"[object RegExp]"===i?(i=new o.constructor(o.source,Oe.exec(o)),i.lastIndex=o.lastIndex,i):"[object Set]"===i?new r:"[object Symbol]"===i?Gn?Object(Gn.call(o)):{}:void 0}function ht(e){return co(e)||_o(e)||!!(jn&&e&&e[jn])}function gt(n,t){var o=dn(n);return t=null==t?9007199254740991:t,!!t&&("number"==o||"symbol"!=o&&Re.test(n))&&-1<n&&0==n%1&&n<t}function ft(o,t,e){if(!Pt(e))return!1;var r=dn(t);return!("number"==r?!(Nt(e)&>(t,e.length)):!("string"==r&&t in e))&&Ct(e[t],o)}function yt(n,t){if(co(n))return!1;var e=dn(n);return"number"==e||"symbol"==e||"boolean"==e||null==n||Mt(n)||de.test(n)||!ve.test(n)||null!=t&&n in Object(t)}function mt(n){var t=n&&n.constructor;return n===("function"==typeof t&&t.prototype||Qe)}function wt(n,t){return function(e){return null!=e&&e[n]===t&&(t!==Ce||n in Object(e))}}function bt(t,e,r){return e=On(e===Ce?t.length-1:e,0),function(){for(var n=arguments,o=-1,d=On(n.length-e,0),i=Array(d);++o<d;)i[o]=n[e+o];for(o=-1,d=Array(e+1);++o<e;)d[o]=n[o];return d[e]=r(i),an(t,this,d)}}function kt(i){var t=0,e=0;return function(){var r=Fn(),n=16-(r-e);if(!(e=r,0<n))t=0;else if(800<=++t)return arguments[0];return i.apply(Ce,arguments)}}function xt(n){if("string"==typeof n||Mt(n))return n;var o=n+"";return"0"==o&&1/n==-he?"-0":o}function jt(e){if(null!=e){try{return Ye.call(e)}catch(e){}return e+""}return""}function vt(n,t){return sn(ye,function(r){var e="_."+r[0];t&r[1]&&!pn(n,e)&&n.push(e)}),n.sort()}function St(n){if(n instanceof y)return n.clone();var t=new g(n.__wrapped__,n.__chain__);return t.__actions__=Q(n.__actions__),t.__index__=n.__index__,t.__values__=n.__values__,t}function At(e){return(null==e?0:e.length)?z(e,1):[]}function zt(n){var t=null==n?0:n.length;return t?n[t-1]:Ce}function It(e){return null==e?e:Nn.call(e)}function Ot(e){return e=u(e),e.__chain__=!0,e}function Et(n,t){return t(n)}function Ft(i,d){function a(){var e=arguments,t=d?d.apply(this,e):e[0],n=a.cache;return n.has(t)?n.get(t):(e=i.apply(this,e),a.cache=n.set(t,e)||n,e)}if("function"!=typeof i||null!=d&&"function"!=typeof d)throw new TypeError("Expected a function");
577return a.cache=new(Ft.Cache||f),a}function Rt(n){if("function"!=typeof n)throw new TypeError("Expected a function");return function(){var t=arguments;switch(t.length){case 0:return!n.call(this);case 1:return!n.call(this,t[0]);case 2:return!n.call(this,t[0],t[1]);case 3:return!n.call(this,t[0],t[1],t[2]);}return!n.apply(this,t)}}function Ct(n,t){return n===t||n!==n&&t!==t}function Nt(e){return null!=e&&Tt(e.length)&&!Wt(e)}function Lt(e){return Bt(e)&&Nt(e)}function Wt(e){return!!Pt(e)&&(e=R(e),"[object Function]"==e||"[object GeneratorFunction]"==e||"[object AsyncFunction]"==e||"[object Proxy]"==e)}function Tt(e){return"number"==typeof e&&-1<e&&0==e%1&&9007199254740991>=e}function Pt(n){var t=dn(n);return null!=n&&("object"==t||"function"==t)}function Bt(e){return null!=e&&"object"==dn(e)}function Vt(e){return"string"==typeof e||!co(e)&&Bt(e)&&"[object String]"==R(e)}function Mt(e){return"symbol"==dn(e)||Bt(e)&&"[object Symbol]"==R(e)}function qt(n){if(!n)return[];if(Nt(n))return Vt(n)?Ie.test(n)?n.match(Me)||[]:n.split(""):Q(n);if(vn&&n[vn]){n=n[vn]();for(var o=[],r;!(r=n.next()).done;)o.push(r.value);return o}return r=ro(n),("[object Map]"==r?l:"[object Set]"==r?p:oe)(n)}function $t(e){return e?(e=Kt(e),e===he||e===-he?1.7976931348623157e308*(0>e?-1:1):e===e?e:0):0===e?e:0}function Jt(n){n=$t(n);var t=n%1;return n===n?t?n-t:n:0}function Kt(n){if("number"==typeof n)return n;if(Mt(n))return Ht;if(Pt(n)&&(n="function"==typeof n.valueOf?n.valueOf():n,n=Pt(n)?n+"":n),"string"!=typeof n)return 0===n?n:+n;n=n.replace(je,"");var o=xe.test(n);return o||Ee.test(n)?Ue(n.slice(2),o?2:8):Se.test(n)?Ht:+n}function Gt(e){return null==e?"":$(e)}function te(n,t,e){return n=null==n?Ce:E(n,t),n===Ce?e:n}function ee(d,t){var e;if(e=null!=d){e=d;var r=H(t,e);for(var n=-1,o=r.length,s=!1,i;++n<o&&(i=xt(r[n]),!!(s=null!=e&&null!=e&&i in Object(e)));)e=e[i];s||++n!=o?e=s:(o=null==e?0:e.length,e=!!o&&Tt(o)&>(i,o)&&(co(e)||_o(e)))}return e}function re(n){if(Nt(n))n=b(n);else if(mt(n)){var t=[],o;for(o in Object(n))Ze.call(n,o)&&"constructor"!=o&&t.push(o);n=t}else n=In(n);return n}function ne(o){if(Nt(o))o=b(o,!0);else if(Pt(o)){var t=mt(o),i=[],n;for(n in o)("constructor"!=n||!t&&Ze.call(o,n))&&i.push(n);o=i}else{if(n=[],null!=o)for(t in Object(o))n.push(t);o=n}return o}function oe(e){return null==e?[]:r(e,re(e))}function ue(e){return function(){return e}}function ie(e){return e}function ae(e){return W("function"==typeof e?e:A(e,1))}function ce(d,s,e){var n=re(s),o=O(s,n);null!=e||Pt(s)&&(o.length||!n.length)||(e=s,s=d,d=this,o=O(s,re(s)));var i=!(Pt(e)&&"chain"in e&&!e.chain),r=Wt(d);return sn(o,function(o){var a=s[o];d[o]=a,r&&(d.prototype[o]=function(){var o=this.__chain__;if(i||o){var e=d(this.__wrapped__);return(e.__actions__=Q(this.__actions__)).push({func:a,args:arguments,thisArg:d}),e.__chain__=o,e}return a.apply(d,cn([this.value()],arguments))})}),d}function fe(){}function se(e){return yt(e)?o(xt(e)):D(e)}function _e(){return[]}function le(){return!1}var he=1/0,Ht=NaN,ye=[["ary",128],["bind",1],["bindKey",2],["curry",8],["curryRight",16],["flip",512],["partial",32],["partialRight",64],["rearg",256]],ve=/\.|\[(?:[^[\]]*|(["'])(?:(?!\1)[^\\]|\\.)*?\1)\]/,de=/^\w*$/,ge=/[^.[\]]+|\[(?:(-?\d+(?:\.\d+)?)|(["'])((?:(?!\2)[^\\]|\\.)*?)\2)\]|(?=(?:\.|\[\])(?:\.|\[\]|$))/g,je=/^\s+|\s+$/g,we=/\{(?:\n\/\* \[wrapped with .+\] \*\/)?\n?/,Ae=/\{\n\/\* \[wrapped with (.+)\] \*/,me=/,? & /,ke=/\\(\\)?/g,Oe=/\w*$/,Se=/^[-+]0x[0-9a-f]+$/i,xe=/^0b[01]+$/i,ze=/^\[object .+?Constructor\]$/,Ee=/^0o[0-7]+$/i,Re=/^(?:0|[1-9]\d*)$/,Me=/\ud83c[\udffb-\udfff](?=\ud83c[\udffb-\udfff])|(?:[^\ud800-\udfff][\u0300-\u036f\ufe20-\ufe2f\u20d0-\u20ff]?|[\u0300-\u036f\ufe20-\ufe2f\u20d0-\u20ff]|(?:\ud83c[\udde6-\uddff]){2}|[\ud800-\udbff][\udc00-\udfff]|[\ud800-\udfff])[\ufe0e\ufe0f]?(?:[\u0300-\u036f\ufe20-\ufe2f\u20d0-\u20ff]|\ud83c[\udffb-\udfff])?(?:\u200d(?:[^\ud800-\udfff]|(?:\ud83c[\udde6-\uddff]){2}|[\ud800-\udbff][\udc00-\udfff])[\ufe0e\ufe0f]?(?:[\u0300-\u036f\ufe20-\ufe2f\u20d0-\u20ff]|\ud83c[\udffb-\udfff])?)*/g,Ie=/[\u200d\ud800-\udfff\u0300-\u036f\ufe20-\ufe2f\u20d0-\u20ff\ufe0e\ufe0f]/,Be={},Ce;
577Be["[object Float32Array]"]=Be["[object Float64Array]"]=Be["[object Int8Array]"]=Be["[object Int16Array]"]=Be["[object Int32Array]"]=Be["[object Uint8Array]"]=Be["[object Uint8ClampedArray]"]=Be["[object Uint16Array]"]=Be["[object Uint32Array]"]=!0,Be["[object Arguments]"]=Be["[object Array]"]=Be["[object ArrayBuffer]"]=Be["[object Boolean]"]=Be["[object DataView]"]=Be["[object Date]"]=Be["[object Error]"]=Be["[object Function]"]=Be["[object Map]"]=Be["[object Number]"]=Be["[object Object]"]=Be["[object RegExp]"]=Be["[object Set]"]=Be["[object String]"]=Be["[object WeakMap]"]=!1;var Zt={};Zt["[object Arguments]"]=Zt["[object Array]"]=Zt["[object ArrayBuffer]"]=Zt["[object DataView]"]=Zt["[object Boolean]"]=Zt["[object Date]"]=Zt["[object Float32Array]"]=Zt["[object Float64Array]"]=Zt["[object Int8Array]"]=Zt["[object Int16Array]"]=Zt["[object Int32Array]"]=Zt["[object Map]"]=Zt["[object Number]"]=Zt["[object Object]"]=Zt["[object RegExp]"]=Zt["[object Set]"]=Zt["[object String]"]=Zt["[object Symbol]"]=Zt["[object Uint8Array]"]=Zt["[object Uint8ClampedArray]"]=Zt["[object Uint16Array]"]=Zt["[object Uint32Array]"]=!0,Zt["[object Error]"]=Zt["[object Function]"]=Zt["[object WeakMap]"]=!1;var Ue=parseInt,Le="object"==("undefined"==typeof Dt?"undefined":dn(Dt))&&Dt&&Dt.Object===Object&&Dt,Pe="object"==("undefined"==typeof self?"undefined":dn(self))&&self&&self.Object===Object&&self,Ne=Le||Pe||Function("return this")(),Ve="object"==("undefined"==typeof $e?"undefined":dn($e))&&$e&&!$e.nodeType&&$e,We=Ve&&"object"==("undefined"==typeof Fe?"undefined":dn(Fe))&&Fe&&!Fe.nodeType&&Fe,Te=We&&We.exports===Ve,Ge=Te&&Le.process,Ke;t:{try{Ke=Ge&&Ge.binding&&Ge.binding("util");break t}catch(e){}Ke=void 0}var De=Ke&&Ke.isMap,qe=Ke&&Ke.isSet,Je=Ke&&Ke.isTypedArray,He=Array.prototype,Qe=Object.prototype,Xe=Ne["__core-js_shared__"],Ye=Function.prototype.toString,Ze=Qe.hasOwnProperty,Yt=function(){var e=/[^.]+$/.exec(Xe&&Xe.keys&&Xe.keys.IE_PROTO||"");return e?"Symbol(src)_1."+e:""}(),Qt=Qe.toString,Xt=RegExp("^"+Ye.call(Ze).replace(/[\\^$.*+?()[\]{}|]/g,"\\$&").replace(/hasOwnProperty|(function).*?(?=\\\()| for .+?(?=\\\])/g,"$1.*?")+"$"),gn=Te?Ne.Buffer:Ce,fn=Ne.Symbol,yn=Ne.Uint8Array,mn=gn?gn.f:Ce,wn=t(Object.getPrototypeOf),bn=Object.create,kn=Qe.propertyIsEnumerable,xn=He.splice,jn=fn?fn.isConcatSpreadable:Ce,vn=fn?fn.iterator:Ce,Sn=fn?fn.toStringTag:Ce,An=function(){try{var e=ct(Object,"defineProperty");return e({},"",{}),e}catch(e){}}(),Tn=Object.getOwnPropertySymbols,zn=gn?gn.isBuffer:Ce,In=t(Object.keys),On=T,En=V,Fn=Date.now,Nn=He.reverse,Rn=ct(Ne,"DataView"),Wn=ct(Ne,"Map"),Cn=ct(Ne,"Promise"),Ln=ct(Ne,"Set"),Pn=ct(Ne,"WeakMap"),Bn=ct(Object,"create"),Un=Pn&&new Pn,Vn={},Mn=jt(Rn),Dn=jt(Wn),qn=jt(Cn),$n=jt(Ln),Jn=jt(Pn),Kn=fn?fn.prototype:Ce,Gn=Kn?Kn.valueOf:Ce,Hn=Kn?Kn.toString:Ce,Zn=function(){function n(){}return function(t){return Pt(t)?bn?bn(t):(n.prototype=t,t=new n,n.prototype=Ce,t):{}}}();u.prototype=d.prototype,u.prototype.constructor=u,g.prototype=Zn(d.prototype),g.prototype.constructor=g,y.prototype=Zn(d.prototype),y.prototype.constructor=y,c.prototype.clear=function(){this.__data__=Bn?Bn(null):{},this.size=0},c.prototype["delete"]=function(e){return e=this.has(e)&&delete this.__data__[e],this.size-=e?1:0,e},c.prototype.get=function(n){var t=this.__data__;return Bn?(n=t[n],"__lodash_hash_undefined__"===n?Ce:n):Ze.call(t,n)?t[n]:Ce},c.prototype.has=function(n){var t=this.__data__;return Bn?t[n]!==Ce:Ze.call(t,n)},c.prototype.set=function(n,t){var e=this.__data__;return this.size+=this.has(n)?0:1,e[n]=Bn&&t===Ce?"__lodash_hash_undefined__":t,this},h.prototype.clear=function(){this.__data__=[],this.size=0},h.prototype["delete"]=function(n){var t=this.__data__;return n=x(t,n),!(0>n)&&(n==t.length-1?t.pop():xn.call(t,n,1),--this.size,!0)},h.prototype.get=function(n){var t=this.__data__;return n=x(t,n),0>n?Ce:t[n][1]},h.prototype.has=function(e){return-1<x(this.__data__,e)},h.prototype.set=function(o,t){var e=this.__data__,r=x(e,o);return 0>r?(++this.size,e.push([o,t])):e[r][1]=t,this},f.prototype.clear=function(){this.size=0,this.__data__={hash:new c,map:new(Wn||h),string:new c}},f.prototype["delete"]=function(e){return e=pt(this,e)["delete"](e),this.size-=e?1:0,e},f.prototype.get=function(e){return pt(this,e).get(e)},f.prototype.has=function(e){return pt(this,e).has(e)},f.prototype.set=function(o,t){var e=pt(this,o),r=e.size;return e.set(o,t),this.size+=e.size==r?0:1,this},m.prototype.add=m.prototype.push=function(e){return this.__data__.set(e,"__lodash_hash_undefined__"),this},m.prototype.has=function(e){return this.__data__.has(e)},w.prototype.clear=function(){this.__data__=new h,this.size=0},w.prototype["delete"]=function(n){var t=this.__data__;return n=t["delete"](n),this.size=t.size,n},w.prototype.get=function(e){return this.__data__.get(e)},w.prototype.has=function(e){return this.__data__.has(e)},w.prototype.set=function(o,t){var e=this.__data__;if(e instanceof h){var r=e.__data__;if(!Wn||199>r.length)return r.push([o,t]),this.size=++e.size,this;e=this.__data__=new f(r)}return e.set(o,t),this.size=e.size,this};var Yn=function(d,t){return function(e,r){if(null==e)return e;if(!Nt(e))return d(e,r);for(var n=e.length,o=t?n:-1,a=Object(e);(t?o--:++o<n)&&!1!==r(a[o],o,a););return e}}(I),Qn=function(e){return function(t,d,r){var n=-1,o=Object(t);r=r(t);for(var s=r.length,i;s--&&(i=r[e?s:++n],!1!==d(o[i],i,o)););return t}}(),Xn=Un?function(n,t){return Un.set(n,t),n}:ie,eo=An?function(n,t){return An(n,"toString",{configurable:!0,enumerable:!1,value:ue(t),writable:!0})}:ie,to=Un?function(e){return Un.get(e)}:fe,no=Tn?function(n){return null==n?[]:(n=Object(n),ln(Tn(n),function(t){return kn.call(n,t)}))}:_e,oo=Tn?function(n){for(var t=[];n;)cn(t,no(n)),n=wn(n);return t}:_e,ro=R;(Rn&&"[object DataView]"!=ro(new Rn(new ArrayBuffer(1)))||Wn&&"[object Map]"!=ro(new Wn)||Cn&&"[object Promise]"!=ro(Cn.resolve())||Ln&&"[object Set]"!=ro(new Ln)||Pn&&"[object WeakMap]"!=ro(new Pn))&&(ro=function(n){var t=R(n);if(n=(n="[object Object]"==t?n.constructor:Ce)?jt(n):"")switch(n){case Mn:return"[object DataView]";case Dn:return"[object Map]";case qn:return"[object Promise]";case $n:return"[object Set]";case Jn:return"[object WeakMap]";}return t});var io=kt(Xn),ao=kt(eo),so=function(n){n=Ft(n,function(e){return 500===o.size&&o.clear(),e});var o=n.cache;return n}(function(n){var i=[];return 46===n.charCodeAt(0)&&i.push(""),n.replace(ge,function(e,t,r,n){i.push(r?n.replace(ke,"$1"):t||e)}),i}),lo=q(function(n){var t=_n(n,K);if(t.length&&t[0]===n[0]){n=t[0].length;for(var e=t.length,o=e,r=Array(e),i=1/0,d=[];o--;){var a=t[o],i=En(a.length,i);r[o]=120<=n&&120<=a.length?new m(o&&a):Ce}var a=t[0],c=-1,s=r[0];t:for(;++c<n&&d.length<i;){var _=a[c],l=_,_=0===_?0:_;if(s?!s.has(l):!pn(d,l)){for(o=e;--o;){var u=r[o];if(u?!u.has(l):!pn(t[o],l))continue t}s&&s.push(l),d.push(_)}}t=d}else t=[];return t});(function(e){return ao(bt(e,Ce,At),e+"")})(function(i){function t(t){for(var d=-1,r=i.length,n=Array(r);++d<r;)n[d]=null==t?Ce:te(t,i[d]);return n}var e=i.length,r=e?i[0]:0,d=this.__wrapped__;return!(1<e||this.__actions__.length)&&d instanceof y&>(r)?(d=d.slice(r,+r+(e?1:0)),d.__actions__.push({func:Et,args:[t],thisArg:Ce}),new g(d,this.__chain__).thru(function(n){return e&&!n.length&&n.push(Ce),n})):this.thru(t)});var po=q(function(n,t){if(null==n)return[];var e=t.length;return 1<e&&ft(n,t[0],t[1])?t=[]:2<e&&ft(t[0],t[1],t[2])&&(t=[t[0]]),M(n,z(t,1),[])});Ft.Cache=f;var _o=C(function(){return arguments}())?C:function(e){return Bt(e)&&Ze.call(e,"callee")&&!kn.call(e,"callee")},co=Array.isArray,tn=zn||le,en=De?s(De):function(e){return Bt(e)&&"[object Map]"==ro(e)},rn=qe?s(qe):function(e){return Bt(e)&&"[object Set]"==ro(e)},nn=Je?s(Je):function(e){return Bt(e)&&Tt(e.length)&&!!Be[R(e)]},on=function(n){return function(t){var e=ro(t);return"[object Map]"==e?l(t):"[object Set]"==e?_(t):a(t,n(t))}}(re);u.chain=Ot,u.constant=ue,u.flatten=At,u.intersection=lo,u.iteratee=ae,u.keys=re,u.keysIn=ne,u.map=function(n,t){return(co(n)?_n:P)(n,lt(t,3))},u.mapKeys=function(n,i){var e={};return i=lt(i,3),I(n,function(r,t,n){S(e,i(r,t,n),r)}),e},u.mapValues=function(n,i){var e={};return i=lt(i,3),I(n,function(r,t,n){S(e,t,i(r,t,n))}),e},u.memoize=Ft,u.mixin=ce,u.negate=Rt,u.orderBy=function(o,t,e,r){return null==o?[]:(co(t)||(t=null==t?[]:[t]),e=r?Ce:e,co(e)||(e=null==e?[]:[e]),M(o,t,e))},u.property=se,u.reverse=It,u.sortBy=po,u.tap=function(n,t){return t(n),n},u.thru=Et,u.toArray=qt,u.toPairs=on,u.values=oe,u.entries=on,ce(u,u),u.eq=Ct,u.get=te,u.hasIn=ee,u.identity=ie,u.isArguments=_o,u.isArray=co,u.isArrayLike=Nt,u.isArrayLikeObject=Lt,u.isBuffer=tn,u.isFunction=Wt,u.isLength=Tt,u.isMap=en,u.isObject=Pt,u.isObjectLike=Bt,u.isSet=rn,u.isString=Vt,u.isSymbol
577=Mt,u.isTypedArray=nn,u.last=zt,u.stubArray=_e,u.stubFalse=le,u.noop=fe,u.toFinite=$t,u.toInteger=Jt,u.toNumber=Kt,u.toString=Gt,ce(u,function(){var n={};return I(u,function(t,e){Ze.call(u.prototype,e)||(n[e]=t)}),n}(),{chain:!1}),u.VERSION="4.17.5",sn(["drop","take"],function(o,t){y.prototype[o]=function(i){i=i===Ce?1:On(Jt(i),0);var r=this.__filtered__&&!t?new y(this):this.clone();return r.__filtered__?r.__takeCount__=En(i,r.__takeCount__):r.__views__.push({size:En(i,4294967295),type:o+(0>r.__dir__?"Right":"")}),r},y.prototype[o+"Right"]=function(t){return this.reverse()[o](t).reverse()}}),sn(["filter","map","takeWhile"],function(n,t){var o=t+1;y.prototype[n]=function(n){var t=this.clone();return t.__iteratees__.push({iteratee:lt(n,3),type:o}),t.__filtered__=t.__filtered__||1==o||3==o,t}}),sn(["head","last"],function(n,t){var e="take"+(t?"Right":"");y.prototype[n]=function(){return this[e](1).value()[0]}}),sn(["initial","tail"],function(n,t){var e="drop"+(t?"":"Right");y.prototype[n]=function(){return this.__filtered__?new y(this):this[e](1)}}),y.prototype.compact=function(){return this.filter(ie)},y.prototype.find=function(e){return this.filter(e).head()},y.prototype.findLast=function(e){return this.reverse().find(e)},y.prototype.invokeMap=q(function(t,d){return"function"==typeof t?new y(this):this.map(function(e){var n=e;e=t;var o=e=H(e,n);if(!(2>o.length)){var r=0,s=-1,a=-1,l=o.length;for(0>r&&(r=-r>l?0:l+r),s=s>l?l:s,0>s&&(s+=l),l=r>s?0:s-r>>>0,r>>>=0,s=Array(l);++a<l;)s[a]=o[a+r];n=E(n,s)}return e=null==n?n:n[xt(zt(e))],null==e?Ce:an(e,n,d)})}),y.prototype.reject=function(e){return this.filter(Rt(lt(e)))},y.prototype.slice=function(n,o){n=Jt(n);var i=this;return i.__filtered__&&(0<n||0>o)?new y(i):(0>n?i=i.takeRight(-n):n&&(i=i.drop(n)),o!==Ce&&(o=Jt(o),i=0>o?i.dropRight(-o):i.take(o-n)),i)},y.prototype.takeRightWhile=function(e){return this.reverse().takeWhile(e).reverse()},y.prototype.toArray=function(){return this.take(4294967295)},I(y.prototype,function(d,t){var p=/^(?:filter|find|map|reject)|While$/.test(t),r=/^(?:head|last)$/.test(t),e=u[r?"take"+("last"==t?"Right":""):t],n=r||/^find/.test(t);e&&(u.prototype[t]=function(){function t(n){return n=e.apply(u,cn([n],a)),r&&_?n[0]:n}var o=this.__wrapped__,a=r?[1]:arguments,c=o instanceof y,f=a[0],m=c||co(o);m&&p&&"function"==typeof f&&1!=f.length&&(c=m=!1);var _=this.__chain__,w=!!this.__actions__.length,f=n&&!_,c=c&&!w;return!n&&m?(o=c?o:new y(this),o=d.apply(o,a),o.__actions__.push({func:Et,args:[t],thisArg:Ce}),new g(o,_)):f&&c?d.apply(this,a):(o=this.thru(t),f?r?o.value()[0]:o.value():o)})}),sn(["pop","push","shift","sort","splice","unshift"],function(o){var i=He[o],e=/^(?:push|sort|unshift)$/.test(o)?"tap":"thru",r=/^(?:pop|shift)$/.test(o);u.prototype[o]=function(){var n=arguments;if(r&&!this.__chain__){var t=this.value();return i.apply(co(t)?t:[],n)}return this[e](function(e){return i.apply(co(e)?e:[],n)})}}),I(y.prototype,function(o,t){var i=u[t];if(i){var d=i.name+"";(Vn[d]||(Vn[d]=[])).push({name:t,func:i})}}),Vn[ot(Ce,2).name]=[{name:"wrapper",func:Ce}],y.prototype.clone=function(){var e=new y(this.__wrapped__);return e.__actions__=Q(this.__actions__),e.__dir__=this.__dir__,e.__filtered__=this.__filtered__,e.__iteratees__=Q(this.__iteratees__),e.__takeCount__=this.__takeCount__,e.__views__=Q(this.__views__),e},y.prototype.reverse=function(){if(this.__filtered__){var e=new y(this);e.__dir__=-1,e.__filtered__=!0}else e=this.clone(),e.__dir__*=-1;return e},y.prototype.value=function(){var d=this.__wrapped__.value(),e=this.__dir__,r=co(d),p=0>e,h=r?d.length:0,u;u=h;for(var g=this.__views__,i=0,y=-1,c=g.length;++y<c;){var f=g[y],s=f.size;switch(f.type){case"drop":i+=s;break;case"dropRight":u-=s;break;case"take":u=En(u,i+s);break;case"takeRight":i=On(i,u-s);}}if(u={start:i,end:u},g=u.start,i=u.end,u=i-g,g=p?i:g-1,i=this.__iteratees__,y=i.length,c=0,f=En(u,this.__takeCount__),!r||!p&&h==u&&f==u)return J(d,this.__actions__);r=[];t:for(;u--&&c<f;){for(g+=e,p=-1,h=d[g];++p<y;){var _=i[p],s=_.type,_=(0,_.iteratee)(h);if(2==s)h=_;else if(!_){if(1==s)continue t;break t}}r[c++]=h}return r},u.prototype.chain=function(){return Ot(this)},u.prototype.commit=function(){return new g(this.value(),this.__chain__)},u.prototype.next=function(){this.__values__===Ce&&(this.__values__=qt(this.value()));var e=this.__index__>=this.__values__.length;return{done:e,value:e?Ce:this.__values__[this.__index__++]}},u.prototype.plant=function(i){for(var t=this,r,e;t instanceof d;){e=St(t),e.__index__=0,e.__values__=Ce,r?n.__wrapped__=e:r=e;var n=e,t=t.__wrapped__}return n.__wrapped__=i,r},u.prototype.reverse=function(){var e=this.__wrapped__;return e instanceof y?(this.__actions__.length&&(e=new y(this)),e=e.reverse(),e.__actions__.push({func:Et,args:[It],thisArg:Ce}),new g(e,this.__chain__)):this.thru(It)}
577,u.prototype.toJSON=u.prototype.valueOf=u.prototype.value=function(){return J(this.__wrapped__,this.__actions__)},u.prototype.first=u.prototype.head,vn&&(u.prototype[vn]=function(){return this}),"function"==typeof G&&"object"==dn(G.amd)&&G.amd?(Ne._=u,G(function(){return u})):We?((We.exports=u)._=u,Ve._=u):Ne._=u}).call(void 0)}).call(this)}).call(this,"undefined"==typeof global?"undefined"==typeof self?"undefined"==typeof window?{}:window:self:global)},{}]},{},[1])(1)}); 578 (function() { 579 var pubElems = document.querySelectorAll(".publication"); 580 var yearElems = document.querySelectorAll(".year"); 581 582 var clearElem = document.getElementById("clear-filters"); 583 var highlightElem = document.getElementById("highlight"); 584 585 var data = []; 586 var allYears = new Set(); 587 588 pubElems.forEach(function(element) { 589 var item = JSON.parse(element.getAttribute("data-pub")); 590 591 if (item.highlight) { 592 item.highlight = "Yes"; 593 } 594 595 allYears.add(item.year); 596 597 item.element = element; 598 599 item.authors = (item.authors || []).filter(function(d) { 600 return d !== "Dominik Moritz"; 601 }); 602 603 data.push(item); 604 }); 605 606 var engine = itemsjs(data, { 607 aggregations: { 608 venue_tags: { 609 size: 5 610 }, 611 authors: { 612 size: 6 613 }, 614 awards: { 615 size: 5 616 }, 617 highlight: { 618 size: 1 619 }, 620 tags: { 621 size: 6 622 }, 623 type: { 624 size: 5 625 } 626 }, 627 searchableFields: ["authors", "awards", "tags", "type", "title", "content"] 628 }); 629 630 // get default search from URL 631 var hash = decodeURIComponent(window.location.hash.substr(1)); 632 633 var result = hash.split('&').reduce(function (res, item) { 634 var [key, value] = item.split('='); 635 if (key && value) { 636 if (key in res) { 637 res[key].push(value) 638 } else { 639 res[key] = [value]; 640 } 641 } 642 return res; 643 }, {}); 644 645 var query = { filters: result }; 646 647 function setAggs(aggs) { 648 document.querySelectorAll("#facets > .facet").forEach(function(facet) { 649 var id = facet.getAttribute("id"); 650 651 var buckets = aggs[id].buckets; 652 653 var el = facet.querySelector("ul"); 654 if (buckets.length === 0) { 655 el.innerHTML = "Empty"; 656 } else { 657 el.innerHTML = ""; 658 659 buckets.forEach(function(bucket) { 660 if (query.filters[id] && query.filters[id].indexOf(bucket.key) >= 0) { 661 bucket.in_query = true; 662 } 663 }); 664 665 var maxDocCount = Math.max.apply( 666 null, 667 buckets.map(function(bucket) { 668 return bucket.doc_count; 669 }) 670 ); 671 672 buckets.forEach(function(bucket) { 673 var child = document.createElement("li"); 674 675 var wrap = document.createElement("span"); 676 child.appendChild(wrap); 677 678 var text = document.createElement("span"); 679 text.classList.add("limited"); 680 text.innerText = bucket.key; 681 text.setAttribute("title", bucket.key); 682 var number = document.createElement("span"); 683 number.classList.add("cnt"); 684 number.innerText = " (" + bucket.doc_count + ")"; 685 wrap.appendChild(text); 686 wrap.appendChild(number); 687 688 var barFull = document.createElement("div"); 689 barFull.classList.add("bar-full"); 690 child.append(barFull); 691 692 var bar = document.createElement("div"); 693 bar.classList.add("bar"); 694 bar.style.width = "" + (bucket.doc_count / maxDocCount) * 100 + "%"; 695 barFull.append(bar); 696 697 if (bucket.in_query) { 698 child.classList.add("in-query"); 699 700 // remove filter 701 child.onclick = function() { 702 query.filters[id].splice( 703 query.filters[id].indexOf(bucket.key), 704 1 705 ); 706 if (query.filters[id].length === 0) { 707 delete query.filters[id]; 708 } 709 search(query); 710 }; 711 } else { 712 // add to filter 713 child.onclick = function() { 714 if (query.filters[id]) { 715 query.filters[id].push(bucket.key); 716 } else { 717 query.filters[id] = [bucket.key]; 718 } 719 search(query); 720 }; 721 } 722 723 el.appendChild(child); 724 }); 725 } 726 }); 727 } 728 729 var ftSearch = document.getElementById("ft-search"); 730 ftSearch.oninput = function() { 731 var val = ftSearch.value; 732 733 if (val) { 734 query.query = val; 735 } else { 736 delete query.query; 737 } 738 739 search(query); 740 } 741 742 function search(query) { 743 console.time("Search"); 744 745 var result = engine.search(Object.assign({ per_page: data.length }, query)); 746 747 setAggs(result.data.aggregations); 748 749 var counter = pubElems.length - result.data.items.length; 750 751 document.getElementById("count_hidden").innerText = counter; 752 document.getElementById("count_total").innerText = pubElems.length; 753
754 pubElems.forEach(function(element) { 755 element.classList.add("hidden"); 756 }); 757 758 var visibleYears = {}; 759 result.data.items.forEach(function(item) { 760 item.element.classList.remove("hidden"); 761 visibleYears[item.year] = 1; 762 }); 763 764 yearElems.forEach(function(element) { 765 element.classList.add("hidden"); 766 }); 767 allYears.forEach(function(year) { 768 if (year in visibleYears) { 769 document.getElementById("y" + year).classList.remove("hidden"); 770 } 771 }); 772 773 // show or hide notification about filtered papers 774 if (Object.keys(query.filters).length || query.query) { 775 clearElem.classList.remove("hidden"); 776 } else { 777 clearElem.classList.add("hidden"); 778 } 779 780 highlightElem.checked = !!query.filters.highlight; 781 782 console.timeEnd("Search"); 783 } 784 785 highlightElem.onchange = function() { 786 if (highlightElem.checked) { 787 query.filters.highlight = ["Yes"]; 788 } else { 789 delete query.filters.highlight; 790 } 791 search(query); 792 }; 793 794 clearElem.onclick = function() { 795 query = { filters: {} }; 796 ftSearch.value = ""; 797 search(query); 798 }; 799 800 search(query); 801 802 document.getElementById("facets").classList.remove("hidden"); 803 document.getElementById("only-highlight").classList.remove("hidden"); 804})(); 805 806</script>
806 807 808 809</section> 810 811 </div> 812 813 <footer> 814 <div class="container"> 815 <div class="footer-col"> 816 Hello from Guangzhou.<br> 817 <abbr title="Last build on 2026-09-11">September 2026</abbr> 818 </div> 819 <div class="footer-col site-desc"> 820 The PhD from HKUST(GZ) working on immersive visualization. I'm <a href="https://twitter.com/zoeywuwa">@zoeywuwa</a> on Twitter and <a rel="me" href="https://www.instagram.com/zoeywang071/">@zoeywang071</a> on Instagram. 821 </div> 822 <div class="footer-col">template from <a href="https://github.com/domoritz/domoritz.github.io">@domoritz</a> 823</div> 824 </div> 825</footer> 826 827
827<script> 828 function trim(str) { 829 return str.replace(/^\s+|\s+$/g, ''); 830 } 831 var headers = document.querySelectorAll("h2, h3, h4, h5, h6"); 832 for (var i=0; i<headers.length; i++) { 833 var h = headers[i]; 834 var name = h.getAttribute("id"); 835 var title = h.innerHTML; 836 h.innerHTML = '<a href="#' + name + '" class="anchor"><i class="fas fa-hashtag"></i></a>' + trim(title); 837 } 838</script>
838 839 840 </body> 841</html>
Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.