https://jiayangao.github.io/

1<!doctype html>
2<html lang="en" class="no-js">
3  <head>
4    <meta charset="utf-8">
5
6<!-- begin SEO -->
7
8
9
10
11
12<title>Jiayang Ao Home</title>
13
14
15
16
17
18
19
20<meta property="og:locale" content="en">
21<meta property="og:site_name" content="Jiayang Ao">
22<meta property="og:title" content="Jiayang Ao">
23
24
25  <link rel="canonical" href="https://jiayangao.github.io/">
26  <meta property="og:url" content="https://jiayangao.github.io/">
27
28
29
30  <meta property="og:description" content="Computer Vision">
31
32
33
34
35
36
37
38  <meta name="google-site-verification" content="fQZQD0EwLaSOrp3EbALt2rjxHEY9pUvWuh1fS96xgNk" />
39
40
41  <meta name="msvalidate.01" content="56D1B3DBCF345410DA9431AFE01788E7">
42
43
44<!-- end SEO -->
45
46<meta name="google-site-verification" content="fQZQD0EwLaSOrp3EbALt2rjxHEY9pUvWuh1fS96xgNk" />
47<meta name="HandheldFriendly" content="True">
48<meta name="MobileOptimized" content="320">
49<meta name="viewport" content="width=device-width, initial-scale=1.0">
50

51<script>
52  document.documentElement.className = document.documentElement.className.replace(/\bno-js\b/g, '') + ' js ';
53</script>

53
54
55<!-- For all browsers -->
56<link rel="stylesheet" href="assets/css/main.css">
57
58<meta http-equiv="cleartype" content="on">
59<head>
60  <base target="_blank">
61</head>
62    <link rel="apple-touch-icon" sizes="180x180" href="images/apple-touch-icon.png">
63<link rel="icon" type="image/png" sizes="32x32" href="images/favicon-32x32.png">
64<link rel="icon" type="image/png" sizes="16x16" href="images/favicon-16x16.png">
65<link rel="manifest" href="images/site.webmanifest">
66
67<meta name="msapplication-TileColor" content="#000000">
68<meta name="msapplication-TileImage" content="images/mstile-144x144.png?v=M44lzPylqQ">
69<meta name="msapplication-config" content="images/browserconfig.xml?v=M44lzPylqQ">
70<meta name="theme-color" content="#ffffff">
71<link rel="stylesheet" href="assets/css/academicons.css"/>
72

73<script type="text/x-mathjax-config"> MathJax.Hub.Config({ TeX: { equationNumbers: { autoNumber: "all" } } }); </script>

73

74<script type="text/x-mathjax-config">
75  MathJax.Hub.Config({
76    tex2jax: {
77      inlineMath: [ ['$','$'], ["\\(","\\)"] ],
78      processEscapes: true
79    }
80  });
81</script>

81

82<script src='https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.4/latest.js?config=TeX-MML-AM_CHTML' async></script>

82
83
84<!-- end custom head snippets -->
85
86  </head>
87
88  <body>
89
90    <!--[if lt IE 9]>
91<div class="notice--danger align-center" style="margin: 0;">You are using an <strong>outdated</strong> browser. Please <a href="http://browsehappy.com/">upgrade your browser</a> to improve your experience.</div>
92<![endif]-->
93    <div class="masthead">
94  <div class="masthead__inner-wrap">
95    <div class="masthead__menu">
96      <nav id="site-nav" class="greedy-nav">
97        <button><div class="navicon"></div></button>
98        <ul class="visible-links">
99          <li class="masthead__menu-item masthead__menu-item--lg masthead__menu-home-item">
100            <a href="#about-me">Homepage</a>
101          </li>
102          
103            <li class="masthead__menu-item"><a href="#about-me">About Me</a></li>
104          
105            <li class="masthead__menu-item"><a href="#-publications">Publications</a></li>
106          
107            <li class="masthead__menu-item"><a href="#-teaching">Teaching</a></li>
108          
109            <li class="masthead__menu-item"><a href="#-awards">Awards</a></li>
110          
111            <li class="masthead__menu-item"><a href="#-education">Education</a></li>
112          
113        </ul>
114        <ul class="hidden-links hidden"></ul>
115      </nav>
116    </div>
117  </div>
118</div>
119
120
121    <div id="main" role="main">
122      
123  <div class="sidebar sticky">
124  
125
126<div itemscope itemtype="http://schema.org/Person" class="profile_box">
127
128  <div class="author__avatar">
129    <img src="images/avatar.jpg" class="author__avatar" alt="Jiayang Ao">
130  </div>
131
132  <div class="author__content">
133    <h3 class="author__name">Jiayang Ao</h3>
134    <p class="author__bio">The University of Melbourne</p>
135  </div>
136
137  <div class="author__urls-wrapper">
138    <!-- <button class="btn btn--inverse">More Info & Contact</button> -->
139    <ul class="author__urls social-icons">
140      
141        <li><div style="white-space: normal; margin-bottom: 1em;">Computer Vision</div></li>
142      
143      
144        <li><i class="fa fa-fw fa-map-marker" aria-hidden="true"></i> Melbourne, Australia</li>
145      
146      
147      
148      
149        <li><a href="mailto:[email protected]"><i class="fas fa-fw fa-envelope" aria-hidden="true"></i> Email</a></li>
150      
151      
152       
153      
154      
155      
156      
157        <li><a href="https://www.linkedin.com/in/jiayang-ao"><i class="fab fa-fw fa-linkedin" aria-hidden="true"></i> LinkedIn</a></li>
158      
159      
160      
161      
162      
163      
164      
165        <li><a href="https://github.com/saraao"><i class="fab fa-fw fa-github" aria-hidden="true"></i> Github</a></li>
166      
167      
168      
169      
170      
171      
172      
173      
174      
175      
176      
177      
178      
179      
180        <li><a href="https://scholar.google.com/citations?user=Sq5U8MAAAAAJ"><i class="fas fa-fw fa-graduation-cap"></i> Google Scholar</a></li>
181      
182      
183      
184        <li><a href="https://orcid.org/0000-0001-6416-1215"><i class="ai ai-orcid-square ai-fw"></i> ORCID</a></li>
185      
186      
187      
188    </ul>
189      <div class="author__urls_sm">
190      
191      
192        <a href="mailto:[email protected]"><i class="fas fa-fw fa-envelope" aria-hidden="true"></i></a>
193      
194      
195       
196      
197      
198      
199      
200        <a href="https://www.linkedin.com/in/jiayang-ao"><i class="fab fa-fw fa-linkedin" aria-hidden="true"></i></a>
201      
202      
203      
204      
205      
206      
207      
208        <a href="https://github.com/saraao"><i class="fab fa-fw fa-github" aria-hidden="true"></i></a>
209      
210      
211      
212      
213      
214      
215      
216      
217      
218      
219      
220      
221      
222      
223        <a href="https://scholar.google.com/citations?user=Sq5U8MAAAAAJ"><i class="fas fa-fw fa-graduation-cap"></i></a>
224      
225      
226      
227        <a href="https://orcid.org/0000-0001-6416-1215"><i class="ai ai-orcid-square ai-fw"></i></a>
228      
229      
230      
231    </div>
232  </div>
233</div>
234
235  
236  </div>
237
238    
239      <article class="page" itemscope itemtype="http://schema.org/CreativeWork">
240        <meta itemprop="headline" content="">
241        <div class="page__inner-wrap">
242          <section class="page__content" itemprop="text">
243            
244<p><span class="anchor" id="about-me"></span></p>
245<h1 id="about-me">About Me</h1>
246<p>

246I am a Research Scientist at Canva CORE Research, working on computer vision and generative AI. My research broadly focuses on visual understanding and generation, particularly on how models understand, represent, and manipulate scenes and objects. My expertise includes generative image editing, image decomposition, amodal perception, structured visual representations, and controllable visual generation.</p>
247
248<p>My work spans inference-time algorithms for visual understanding and generation, vision and generative models for downstream understanding and manipulation tasks, and the creation of high-quality structured visual data with supervision that is difficult or impossible to obtain from real-world data alone. This includes developing methods to improve the quality, controllability, and consistency of image generation and editing, building models for visual understanding and manipulation, and creating rich, structured visual supervision from natural images and fully controllable 3D environments. My research at Canva has contributed to multiple production capabilities and products.</p>
249
250<p>I also actively collaborate with academic researchers on visual reasoning, multimodal learning, 3D/4D scene understanding and spatial intelligence, robot interaction, out-of-distribution detection, and visual reprogramming. <strong>I am open to research collaborations.</strong></p>
251
252<p>I received my PhD in Computer Vision from the University of Melbourne in 2025, supervised by <a href="http://www.kehinger.com/">Associate Professor Krista A. Ehinger</a> and <a href="https://research.monash.edu/en/persons/qiuhong-ke">Dr Qiuhong Ke</a>. My doctoral research focused on visual amodal completion, with an emphasis on understanding and reconstructing objects and scenes beyond directly visible image content.</p>
253
254<p><span class="anchor" id="-publications"></span></p>
255<h1 id="-publications">📝 Publications</h1>
256<ul>
257  <li>Zizhao Li, Zhengkang Xiang, <code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Joseph West, Kourosh Khoshelham. <a href="https://www.sciencedirect.com/science/article/pii/S0924271626002984">Relative Energy Learning for LiDAR Out-of-Distribution Detection</a>. ISPRS Journal of Photogrammetry and Remote Sensing 239, 334-346, 2026.</li>
258  <li>Zizhao Li, Zhengkang Xiang, <code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Feng Liu, Joseph West, Kourosh Khoshelham.  <a href="https://cs-lzz.github.io/ndp-demo/docs/">Neural Distribution Prior for LiDAR Out-of-Distribution Detection</a>. 🌟CVPR Compute Gold Star. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2026.</li>
259  <li>Yunze Liu, Zifan Wang, Peiran Wu, <code class="language-plaintext highlighter-rouge">Jiayang Ao</code>. <a href="https://www.arxiv.org/abs/2512.01383">PointNet4D: A Lightweight 4D Point Cloud Video Backbone for Online and Offline Perception in Robotic Applications</a>. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2026.</li>
260  <li>Lin Li, Youyang Qu, <code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Ming Ding, Chao Chen, Jun Zhang. Poster: <a href="https://dl.acm.org/doi/abs/10.1145/3719027.3760742">The Art of Deception: Crafting Chimera Images for Covert and Robust Semantic Poisoning Attacks</a>. In Proceedings of the ACM Conference on Computer and Communications Security (CCS) 2025.</li>
261  <li>Yanbei Jiang, Yihao Ding, Chao Lei, <code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Jey Han Lau, Krista A. Ehinger. <a href="https://arxiv.org/abs/2505.21850">Beyond Perception: Evaluating Abstract Visual Reasoning through Multi-Stage Task</a>. In Proceedings of the Association for Computational Linguistics (ACL) 2025 Findings.</li>
262  <li><code class="language-plaintext highlighter-rouge">Jiayang Ao</code>. <a href="https://minerva-access.unimelb.edu.au/items/8f4d1a4b-c56a-4d7c-8ecd-ee1113dece9a">Visual amodal completion</a>. PhD thesis. 2025.</li>
263  <li><code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Yanbei Jiang, Qiuhong Ke, and Krista A Ehinger. <a href="https://arxiv.org/abs/2411.13019">Open-World Amodal Appearance Completion</a>. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2025.</li>
264  <li><code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Qiuhong Ke, and Krista A Ehinger. <a href="https://bmvc2024.org/proceedings/15/">Sequential Amodal Segmentation via Cumulative Occlusion Learning</a>. In Proceedings of the 35th British Machine Vision Conference (BMVC), November 2024.</li>
265  <li><code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Qiuhong Ke, and Krista A. Ehinger. <a href="https://openaccess.thecvf.com/content/WACV2024/html/Ao_Amodal_Intra-Class_Instance_Segmentation_Synthetic_Datasets_and_Benchmark_WACV_2024_paper.html">Amodal intra-class instance segmentation: Synthetic datasets and benchmark</a>. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pages 281–290, January 2024.</li>
266  <li><code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Qiuhong Ke, and Krista A Ehinger. <a href="https://www.sciencedirect.com/science/article/abs/pii/S1077314223000413">Image amodal completion: A survey. Computer Vision and Image Understanding</a>, page 103661, 2023.</li>
267</ul>
268
269<p><span class="anchor" id="-publications"></span></p>
270<h1 id="️-manuscripts">✍️ Manuscripts</h1>
271<ul>
272  <li>Jizhizi Li, <code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Danny Wicks, Petru-Daniel Tudosiu. <a href="https://arxiv.org/abs/2605.14399">SceneForge: Structured World Supervision from 3D Interventions
273</a>. arXiv, 2026.</li>
274  <li>Xueqi Ma, Shuo Yang, Yanbei Jiang, Shu Liu, Zhenzhen Liu, <code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Xingjun Ma, Sarah Monazam Erfani, James Bailey. <a href="https://arxiv.org/abs/2603.20662">Attention in Space: Functional Roles of VLM Heads for Spatial Reasoning</a>. arXiv, 2026.</li>
275  <li>Hongxing Fan, Shuyu Zhao, <code class="language-plaintext highlighter-rouge">Jiayang Ao</code>, Lu Sheng. <a href="https://arxiv.org/abs/2512.20936">Reasoning-Driven Amodal Completion: Collaborative Agents and Perceptual Evaluation</a>. arXiv, 2025.</li>
276</ul>
277
278<p><span class="anchor" id="-appointments"></span></p>
279<h1 id="-professional-appointments">🔬 Professional Appointments</h1>
280<ul>
281  <li>

281Research Scientist, Canva CORE/Leonardo.Ai, 06/2025 - present</li>
282  <li>Research Assistant (Postdoctoral), <a href="https://www.unimelb.edu.au/mdap/research/2023-collaborations/computational-methods-for-artwork-authentication-using-hyperspectral-imaging">ArtCert</a>, Faculty of Arts, University of Melbourne, 04/2025 – 06/2025</li>
283</ul>
284
285<p><span class="anchor" id="-teaching"></span></p>
286<h1 id="-teaching">🏫 Teaching</h1>
287<ul>
288  <li>Lecturer &amp; Course Developer, ISYS3466 Data-Driven Decision Making: Machine Learning for Business Professionals, the RMIT University, 07/2024- 12/2024</li>
289  <li>Head tutor &amp; Course Developer, <a href="https://handbook.unimelb.edu.au/subjects/comp90086">COMP90086 Computer Vision</a>, the University of Melbourne, 07/2021- 12/2024</li>
290  <li>Head tutor, <a href="https://handbook.unimelb.edu.au/2024/subjects/comp90049">COMP90049 Introduction to Machine Learning</a>, the University of Melbourne, 02/2022- 07/2024</li>
291  <li>Head tutor, <a href="https://handbook.unimelb.edu.au/subjects/comp30027">COMP30027 Machine Learning</a>, the University of Melbourne, 02/2022- 06/2025</li>
292</ul>
293
294<p><span class="anchor" id="-awards"></span></p>
295<h1 id="-scholarships-awards--funding">🏆 Scholarships, Awards &amp; Funding</h1>
296<ul>
297  <li>[Jun 2025] CVPR Travel Support Award</li>
298  <li>[Sep 2024] Human Annotation Fund, AI group, School of Computing and Information Systems, the University of Melbourne</li>
299  <li>[Jan 2024] IEEE/CVF Winter Conference on Applications of Computer Vision Travel Grant</li>
300  <li>[Nov 2023] Engineering and IT Conference Travel Scholarship</li>
301  <li>[Nov 2023] Melbourne School of Engineering Travelling Scholarship (CIS)</li>
302  <li>[Nov 2022] <a href="https://eng.unimelb.edu.au/about/feit-excellence-awards/honour-roll/2022-honour-roll">FEIT Excellence Award in Teaching and Learning for Tutors and Demonstrators</a></li>
303  <li>[Aug 2021] Melbourne Research Scholarship (Fee offset and Allowance)</li>
304  <li>[Jun 2019] Outstanding Graduates of Central South University</li>
305  <li>[May 2018] <a href="https://portfolio.justzht.com/areco/">Central China Hackathon Rank #2</a></li>
306  <li>[May 2018] Innovation and Entrepreneurship Training Program for College Students, China National Projects</li>
307  <li>[Jun 2017] China Top10 in ‘Her Startup’ Empowering Women Entrepreneurs Competition for College Students</li>
308  <li>[Jan 2017] Interdisciplinary Contest In Modeling (Honorable Mention)</li>
309</ul>
310
311<p><span class="anchor" id="-services"></span></p>
312<h1 id="️-services">🏛️ SERVICES</h1>
313<h3 id="journal-reviewer">Journal Reviewer</h3>
314<ul>
315  <li>IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)
316    <h3 id="conference-service">Conference Service</h3>
317  </li>
318  <li>Program Committee Member, AAAI Conference on Artificial Intelligence (AAAI)</li>
319  <li>Reviewer, ACM Multimedia (MM)
320    <h3 id="outreach-and-engagement">Outreach and Engagement</h3>
321  </li>
322  <li>Unimelb Open Day Demonstration 2024
323<span class="anchor" id="-xl"></span></li>
324</ul>
325
326<h1 id="-education">🎓 Education</h1>
327<ul>
328  <li><em>2021.10 - 2025.09</em>, <a href="https://www.unimelb.edu.au/"><img class="svg" src="/images/mel.png" width="20pt" /></a> Doctor of Philosophy - Engineering and IT, The University of Melbourne</li>
329  <li><em>2019.07 - 2021.07</em>, <a href="https://www.unimelb.edu.au/"><img class="svg" src="/images/mel.png" width="20pt" /></a> Master of Science (Computer Science) (with distinction), The University of Melbourne</li>
330  <li><em>2015.09 - 2019.06</em>, <a href="https://www.csu.edu.cn//"><img class="svg" src="/images/csu.png" width="20pt" /></a> Bachelor of Management (Information Management and Information Systems), Central South University
331<span class="anchor" id="-lwzl"></span></li>
332</ul>
333
334
335          </section>
336        </div>
337      </article>
338    </div>
339
340    

340<script src="assets/js/main.min.js"></script>

340
341
342
343
344<!-- Global site tag (gtag.js) - Google Analytics -->

345<script async src="https://www.googletagmanager.com/gtag/js?id="></script>

345

346<script>
347  window.dataLayer = window.dataLayer || [];
348  function gtag(){dataLayer.push(arguments);}
349  gtag('js', new Date());
350
351  gtag('config', "");
352</script>

352
353
354

355<script>
356    $(document).ready(function () {
357        
358        var gsDataBaseUrl = 'https://raw.githubusercontent.com/jiayangao/jiayangao.github.io/'
359        
360        $.getJSON(gsDataBaseUrl + "google-scholar-stats/gs_data.json", function (data) {
361            var totalCitation = data['citedby']
362            document.getElementById('total_cit').innerHTML = totalCitation;
363            var citationEles = document.getElementsByClassName('show_paper_citations')
364            Array.prototype.forEach.call(citationEles, element => {
365                var paperId = element.getAttribute('data')
366                var numCitations = data['publications'][paperId]['num_citations']
367                element.innerHTML = '| Citations: ' + numCitations;
368            });
369        });
370    })
371</script>

371
372
373
374<div class="footer_site">
375<hr/>
376
377</div>
378  </body>
379</html>