1<!DOCTYPE HTML> 2<!-- 3 Strata by HTML5 UP 4 html5up.net | @ajlkn 5 Free for personal and commercial use under the CCA 3.0 license (html5up.net/license) 6--> 7<html> 8 <head> 9 <title>Yue's Homepage</title> 10 <meta charset="utf-8" /> 11 <meta name="viewport" content="width=device-width, initial-scale=1" /> 12 <!--[if lte IE 8]>
12<script src="assets/js/ie/html5shiv.js"></script>
12<![endif]--> 13 <link rel="stylesheet" href="assets/css/main.css" /> 14 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.7.2/css/all.min.css" integrity="sha512-Evv84Mr4kqVGRNSgIGL/F/aIDqQb7xQ2vcrdIwxfjThSH8CSR7PBEakCr51Ck+w+/U6swU2Im1vVX0SVk9ABhg==" crossorigin="anonymous" referrerpolicy="no-referrer" /> 15 <!--[if lte IE 8]><link rel="stylesheet" href="assets/css/ie8.css" /><![endif]--> 16 </head> 17 <body id="top"> 18 19 <!-- Header --> 20 <header id="header"> 21 <div class="inner"> 22 <a href="index.html" class="image avatar"><img src="images/zhaoyue-2017_small.jpg" alt="" /></a> 23 <h1>I am <strong>Yue Zhao</strong>, <br /> 24 a computer visionary.<br /></h1> 25 <!-- >crafted by <a href="http://html5up.net">HTML5 UP</a>.</h1> <!--> 26 </div> 27 </header> 28 29 <!-- Main --> 30 <div id="main"> 31 32 <!-- One --> 33 <section id="one"> 34 <header class="major"> 35 <h2>About me<br /></h2> 36 </header> 37 <p>My name is Yue Zhao<a href="#pronouce">*</a> (赵岳 in simplified Chinese). 38 I am a Postdoctoral Scholar at Stanford University, advised by <a href="https://stanford.edu/~eadeli/">Prof. Ehsan Adeli</a>. 39 I am affiliated with <a href="https://stai.stanford.edu/">Stanford Translational AI (STAI) in Medicine and Mental Health</a> and <a href="https://svl.stanford.edu/">Stanford Vision and Learning Lab</a>. 40 I recently got my PhD from the <a href="https://www.utexas.edu/">University of Texas at Austin</a>, supervised by <a href="https://www.philkr.net/">Prof. Philipp Krähenbühl</a>. 41 I obtained my MPhil's degree from <a href="http://mmlab.ie.cuhk.edu.hk">Multimedia Laboratory</a> at the <a hef="https://www.cuhk.edu.hk/english/index.html">Chinese University of Hong Kong</a>, supervised by <a href="http://dahua.me">Prof. Dahua Lin</a>. 42 More previously, I got my Bachelor's degrees from Tsinghua University. 43 My research interests are in computer vision, with an emphasis on video analysis and understanding. I am currently interested in (1) video compression in pursuit of visual Intelligence, (2) understanding long-form streaming videos, and (3) learning executable actions from videos. 44 I am a recipient of the 2024-2025 NVIDIA Graduate Fellowship. </p> 45 </section> 46 47 <!-- Two --> 48 <section id="two"> 49 <header class="major"> 50 <h2>News<br /></h2> 51 </header> 52 <p> 53 [Dec 16, 2025] One tech report on quantization is on <a href="https://arxiv.org/abs/2512.14697">arXiv</a>. A supplementary blog post is <a href="https://ai.stanford.edu/~yzz/blog/articles/npq.html">here</a>.<br> 54 [Oct 21, 2025] <a href="https://arxiv.org/pdf/2505.00702.pdf">RayZer</a> won ICCV 2025 Best Student Paper Honorable Mention! <br> 55 [Oct 16, 2025] I am selected as a <a href="https://neurips.cc/Conferences/2025/ProgramCommittee">NeurIPS 2025 "Reviewer : Top Reviewer"</a>! <br> 56 [Jan 22, 2025] <a href="https://arxiv.org/pdf/2406.07548.pdf">BSQ</a> and <a href="https://www.biorxiv.org/content/biorxiv/early/2024/11/11/2024.11.08.622579.full.pdf">ISM</a> are accepted to ICLR 2025! <br> 57 [Jun 11, 2024] One tech report on generic visual tokenizer is on <a href="https://arxiv.org/abs/2406.07548">arXiv</a>.<br> 58 [Jun 07, 2024] <a href="https://arxiv.org/pdf/2212.04501.pdf">LaViLa</a> (CVPR 2023) wins an <a href="https://egovis.github.io/awards/2022_2023/">Egocentric Vision (EgoVis) 2022/2023 Distinguished Paper Award</a>!<br> 59 [May 01, 2024] ð³ï¸âð⤠<a href="https://arxiv.org/pdf/2402.13217.pdf">VideoPrism</a> accepted to ICML!<br> 60 [Apr 20, 2024] Our positive-congruent training paper accepted by TPAMI (finally)! <br> 61 [Feb 26, 2024] One paper accepted to CVPR 2024. See you in Seattle this summer!<br> 62 [Feb 20, 2024] One tech report on foundational video encoder is on <a href="https://arxiv.org/abs/2402.13217">arXiv</a>. It is fueled by <a href="https://arxiv.org/abs/2401.06129">VIIT</a>'s captions.<br> 63 [Jan 11, 2024] One tech report on video instruction tuning (VIIT) is available on <a href="https://arxiv.org/abs/2401.06129">arXiv</a>.<br> 64 [Dec 08, 2023] I am awarded the <a href="https://blogs.nvidia.com/blog/graduate-research-fellowships-for-2024/">2024-2025 NVIDIA Graduate Fellowship</a>
64. Thank you NVIDIA!<br> 65 [Jun 19, 2023] We won <a href="https://epic-kitchens.github.io/2023#results">EPIC-Kitchens 2023</a> Action Recognition and Multi-Instance Retrieval Challenges! I gave a talk on the winning solution at the workshop.<br> 66 [Feb 28, 2023] One paper accepted to CVPR 2023 as Highlight. See you in Vancouver this summer!<br> 67 [Aug 07, 2022] One paper accepted to ECCV 2022.<br> 68 [May 16, 2022] One tech report on positive-congruent training is available on <a href="https://arxiv.org/abs/2205.06265">arXiv</a>.<br> 69 [Mar 28, 2022] One paper accepted to CVPR 2022 as Oral.<br> 70 [Aug 20, 2021] Had a wonderful summer at AWS in Seattle.<br> 71 [Mar 09, 2020] Two papers accepted to CVPR 2020 (1 oral + 1 poster).<br> 72 [Aug 02, 2019] The extended version of our ICCV 2017 work has been accepted by IJCV.<br> 73 [Jun 18, 2019] We launch <a href="https://github.com/open-mmlab/mmaction">MMAction</a>, a versatile toolbox for action understanding based on <a href="https://pytorch.org">PyTorch</a>. v0.1.0 is now online! </p> 74 </section> 75 76 <!-- Three --> 77 <section id="three"> 78 <h2>Selected Preprints<br /></h2> 79 <div class="row"> 80 <article class="3u 12u$(xsmall) work-item"> 81 <a><img src="images/publications/npq_venn.png" class="image fit thumb"></a> 82 </article> 83 <article class="9u 12u$(xsmall)"> 84 <h3>Spherical Leech Quantization for Visual Tokenization and Generation<br /></h3> 85 <p> 86 <U>Yue Zhao</U>, Hanwen Jiang, Zhenlin Xu, Chutong Yang, Ehsan Adeli, Philipp Krähenbühl<br /> 87 arXiv:2512.14697 [cs.CV]<br /> 88 [<a href="https://arxiv.org/pdf/2512.14697.pdf">pdf</a>] 89 [<a href="https://ai.stanford.edu/~yzz/blog/articles/npq.html">blog</a>] 90 [<a href="https://x.com/__yuezhao__/status/2002142386609402270?s=20">X</a>] 91 [<a href="https://cs.stanford.edu/~yzz/npq/">project page</a>] 92 [<a href="https://github.com/zhaoyue-zephyrus/InfinityCC">code for generation</a>] 93 [<a href="https://github.com/zhaoyue-zephyrus/bsq-vit">code for compression</a>] 94 </p> 95 </article> 96 </div> 97 <div class="row"> 98 <article class="3u 12u$(xsmall) work-item"> 99 <a><img src="images/publications/qlip.png" class="image fit thumb"></a> 100 </article> 101 <article class="9u 12u$(xsmall)"> 102 <h3>QLIP: Text-Aligned Visual Tokenization Unifies Auto-Regressive Multimodal Understanding and Generation<br /></h3> 103 <p> 104 <U>Yue Zhao</U>, Fuzhao Xue, Scott Reed, Linxi Fan, Yuke Zhu, Jan Kautz, Zhiding Yu, Philipp Krähenbühl, De-An Huang<br /> 105 arXiv:2502.05178 [cs.CV]<br /> 106 [<a href="https://arxiv.org/pdf/2502.05178.pdf">pdf</a>][<a href="https://nvlabs.github.io/QLIP/">project page</a>][<a href="https://github.com/NVlabs/QLIP/tree/main">code</a>] 107 </p> 108 </article> 109 </div> 110 <div class="row"> 111 <article class="3u 12u$(xsmall) work-item"> 112 <a><img src="images/publications/avion.png" class="image fit thumb"></a> 113 </article> 114 <article class="9u 12u$(xsmall)"> 115 <h3>Training a Large Video Model on a Single Machine in a Day<br /></h3> 116 <p> 117 <U>Yue Zhao</U>, Philipp Krähenbühl<br /> 118 arXiv:2309.16669 [cs.CV]<br /> 119 <b style='color:orange'>Winner of the <a href="https://epic-kitchens.github.io/2023#results">EPIC-Kitchens 2023</a> Action Recognition and Multi-Instance Retrieval Challenges at CVPR 2023</b>.<br /> 120 <b style='color:orange'>Invited talk at <a href="https://sites.google.com/view/ego4d-epic-cvpr2023-workshop/">11th EPIC Workshop at CVPR 2023</a></b><br /> 121 [<a href="https://arxiv.org/pdf/2309.16669.pdf">pdf</a>][<a href="https://github.com/zhaoyue-zephyrus/AVION">code</a>][<a href="./files/publications/EPIC_CVPR2023_report.pdf">tech report</a>] 122 </p> 123 </article> 124 </div> 125 <!-- <div class="row"> 126 <article class="3u 12u$(xsmall) work-item"> 127 <a><img src="images/publications/posec3d.png" class="image fit thumb"></a> 128 </article> 129 <article class="9u 12u$(xsmall)"> 130 <h3>Revisiting Skeleton-based Action Recognition<br /></h3> 131 <p> 132 Haodong Duan, <U>Yue Zhao</U>, Kai Chen, Dian Shao, Dahua Lin, Bo Dai<br /> 133 arXiv:2104.13586 [cs.CV]<br /> 134 [<a href="https://arxiv.org/pdf/2104.13586.pdf">pdf</a>] 135 </p> 136 </article> 137 </div> --> 138 <!-- <div class="row"> 139 <article class="3u 12u$(xsmall) work-item"> 140 <a><img src="images/publications/debias.png" class="image fit thumb"></a> 141 </article> 142 <article class="9u 12u$(xsmall)"> 143 <h3>Mitigating Representation Bias in Action Recognition: Algorithms and Benchmarks<br /></h3> 144 <p> 145 Haodong Duan, <U>Yue Zhao</U>, Kai Chen, Yuanjun Xiong, Dahua Lin<br /> 146 ECCV Workshop on AI for Creative Video Editing and Understanding (CVEU), 2022 <br /> 147 [<a href="https://arxiv.org/pdf/2209.09393.pdf">pdf</a>][<a href="https://github.com/kennymckormick/ARAS-Dataset">dataset</a>] 148 </p> 149 </article> 150 </div> --> 151 <!-- <div class="row"> 152 <article class="3u 12u$(xsmall) work-item"> 153 <a><img src="images/publications/activitynet_2017_challenge.png" class="image fit thumb"></a> 154 </article> 155 <article class="9u 12u$(xsmall)"> 156 <h3>CUHK & ETHZ & SIAT Submission to ActivityNet Challenge 2017<br /></h3> 157 <p> 158 <U>Yue Zhao</U>, Bowen Zhang, Zhirong Wu, Shuo Yang, Lei Zhou, Sijie Yan, Limin Wang, Yuanjun Xiong, Yali Wang, Dahua Lin, Yu Qiao, Xiaoou Tang<br /> 159 CVPR Workshop on ActivityNet Large Scale Activity Recognition Challenge, 2017<br /> 160 [<a href="https://arxiv.org/pdf/1710.08011.pdf">pdf</a>] 161 </p> 162 </article> 163 </div> --> 164 <!-- <div class="row"> 165 <article class="3u 12u$(xsmall) work-item"> 166 <a><img src="images/publications/ssn.png" class="image fit thumb"></a> 167 </article> 168 <article class="9u 12u$(xsmall)"> 169 <h3>Temporal Action Detection with Structured Segment Networks<br /></h3> 170 <p> 171 <U>Yue Zhao</U>, Yuanjun Xiong, Limin Wang, Zhirong Wu, Dahua Lin, Xiaoou Tang<br /> 172 arXiv:1704.06228v1 [cs.CV]<br /> 173 [<a href="https://arxiv.org/pdf/1704.06228.pdf">pdf</a>] 174 </p> 175 </article> 176 </div> --> 177 <!-- <div class="row"> 178 <article class="3u 12u$(xsmall) work-item"> 179 <a><img src="images/publications/tdn.png" class="image fit thumb"></a> 180 </article> 181 <article class="9u 12u$(xsmall)"> 182 <h3>A Pursuit of Temporal Accuracy in General Activity Detection<br /></h3> 183 <p> 184 Yuanjun Xiong, <U>Yue Zhao</U>, Limin Wang, Dahua Lin, Xiaoou Tang<br /> 185 arXiv:1703.02716v1 [cs.CV]<br /> 186 [<a href="https://arxiv.org/abs/1703.02716.pdf">pdf</a>] 187 [<a href="https://github.com/yjxiong/action-detection">
187code</a>] 188 </p> 189 </article> 190 </div> --> 191 </section> 192 193 <!-- Four --> 194 <section id="four"> 195 <h2>Selected Publications<br /></h2> 196 <div class="row"> 197 <article class="3u 12u$(xsmall) work-item"> 198 <a><img src="images/publications/ttt.gif" class="image fit thumb"></a> 199 </article> 200 <article class="9u 12u$(xsmall)"> 201 <h3>One-Minute Video Generation with Test-Time Training<br /></h3> 202 <p> 203 Karan Dalal*, Daniel Koceja*, Gashon Hussein*, Jiarui Xu*, <U>Yue Zhao</U>†, Youjin Song†, Shihao Han, Ka Chun Cheung, Jan Kautz, Carlos Guestrin, Tatsunori Hashimoto, Sanmi Koyejo, Yejin Choi, Yu Sun, Xiaolong Wang<br /> 204 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2025<br /> 205 <b style='color:orange'>Invited keynote talk at <a href="https://viscale.github.io/">The Workshop on Test-time Scaling for Computer Vision (ViSCALE) at CVPR2025</a></b><br /> 206 [<a href="https://arxiv.org/pdf/2504.05298.pdf">pdf</a>][<a href="https://test-time-training.github.io/video-dit/">project page</a>][<a href="https://github.com/test-time-training/ttt-video-dit">code</a>] 207 [<a href="files/publications/cvpr25_poster.pdf">poster</a>] 208 </p> 209 </article> 210 <article class="3u 12u$(xsmall) work-item"> 211 <a><img src="images/publications/bsqvit.png" class="image fit thumb"></a> 212 </article> 213 <article class="9u 12u$(xsmall)"> 214 <h3>Image and Video Tokenization with Binary Spherical Quantization<br /></h3> 215 <p> 216 <U>Yue Zhao</U>, Yuanjun Xiong, Philipp Krähenbühl<br /> 217 International Conference on Learning Representations (ICLR), 2025<br /> 218 [<a href="https://arxiv.org/pdf/2406.07548.pdf">pdf</a>][<a href="https://github.com/zhaoyue-zephyrus/bsq-vit">code</a>] 219 [<a href="files/publications/iclr2025_poster.pdf">poster</a>] 220 </p> 221 </article> 222 <article class="3u 12u$(xsmall) work-item"> 223 <a><img src="images/publications/videoprism.png" class="image fit thumb"></a> 224 </article> 225 <article class="9u 12u$(xsmall)"> 226 <h3>VideoPrism: A Foundational Visual Encoder for Video Understanding<br /></h3> 227 <p> 228 Long Zhao*, Nitesh B. Gundavarapu*, Liangzhe Yuan*, Hao Zhou*, ..., <U>Yue Zhao</U>, ..., Mikhail Sirotenko+, Ting Liu+, Boqing Gong+<br /> 229 International Conference on Machine Learning (ICML), 2024<br /> 230 [<a href="https://arxiv.org/pdf/2402.13217.pdf">arXiv</a>] 231 [<a href="https://blog.research.google/2024/02/videoprism-foundational-visual-encoder.html">Blog</a>] 232 </p> 233 </article> 234 <article class="3u 12u$(xsmall) work-item"> 235 <a><img src="images/publications/elodi.png" class="image fit thumb"></a> 236 </article> 237 <article class="9u 12u$(xsmall)"> 238 <h3>ELODI: Ensemble Logit Difference Inhibition for Positive-Congruent Training<br /></h3> 239 <p> 240 <U>Yue Zhao</U>, Yantao Shen, Yuanjun Xiong, Shuo Yang, Wei Xia, Zhuowen Tu, Bernt Schiele, Stefano Soatto<br /> 241 IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI), 2024<br /> 242 [<a href="https://arxiv.org/pdf/2205.06265.pdf">arXiv</a>][<a href="https://github.com/amazon-science/regression-constraint-model-upgrade">code</a>] 243 </p> 244 </article> 245 </div> 246 <div class="row"> 247 <article class="3u 12u$(xsmall) work-item"> 248 <a><img src="images/publications/viit.png" class="image fit thumb"></a> 249 </article> 250 <article class="9u 12u$(xsmall)"> 251 <h3>Distilling Vision-Language Models on Millions of Videos<br /></h3> 252 <p> 253 <U>Yue Zhao</U>, Long Zhao, Xingyi Zhou, Jialin Wu, Chun-Te Chu, Hui Miao, Florian Schroff, Hartwig Adam, Ting Liu, Boqing Gong, Philipp Krähenbühl, Liangzhe Yuan<br /> 254 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2024<br /> 255 [<a href="https://arxiv.org/pdf/2401.06129.pdf">arXiv</a>] 256 [<a href="https://zhaoyue-zephyrus.github.io/video-instruction-tuning">project page</a>] 257 </p> 258 </article> 259 </div> 260 <!-- <div class="row"> 261 <article class="3u 12u$(xsmall) work-item"> 262 <a><img src="images/publications/leap.gif" class="image fit thumb"></a> 263 </article> 264 <article class="9u 12u$(xsmall)"> 265 <h3>LEAP:Liberate Sprase-View 3D Modeling from Camera Poses<br /></h3> 266 <p> 267 Hanwen Jiang, Zhenyu Jiang, <U>Yue Zhao</U>, Qixing Huang<br /> 268 International Conference on Learning Representations (ICLR), 2024<br />
269 [<a href="https://arxiv.org/pdf/2310.01410.pdf">arXiv</a>] 270 [<a href="https://hwjiang1510.github.io/LEAP/">project page</a>] 271 [<a href="https://github.com/hwjiang1510/LEAP">code</a>] 272 </p> 273 </article> 274 </div> --> 275 276 <div class="row"> 277 <article class="3u 12u$(xsmall) work-item"> 278 <a><img src="images/publications/lavila.png" class="image fit thumb"></a> 279 </article> 280 <article class="9u 12u$(xsmall)"> 281 <h3>Learning Video Representations from Large Language Models<br /></h3> 282 <p> 283 <U>Yue Zhao</U>, Ishan Misra, Philipp Krähenbühl, Rohit Girdhar<br /> 284 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2023 285 <b style='color:orange'>(Highlight, top-2.5%)</b><br /> 286 <b style='color:orange'>Invited talk at <a href="https://sites.google.com/view/ego4d-epic-cvpr2023-workshop/">Joint International 3rd Ego4D and 11th EPIC Workshop</a></b><br /> 287 <b style='color:orange'><a href="https://egovis.github.io/awards/2022_2023/"></a>Egocentric Vision (EgoVis) 2022/2023 Distinguished Paper Award</a></b><br /> 288 [<a href="https://arxiv.org/pdf/2212.04501.pdf">arXiv</a>] 289 [<a href="https://github.com/facebookresearch/LaViLa">code</a>] 290 [<a href="https://facebookresearch.github.io/LaViLa/">project page</a>] 291 [<a href="https://huggingface.co/spaces/nateraw/lavila">demo</a>] 292 [<a href="https://colab.research.google.com/drive/1gHWiEWywIotRivYQTR-8NQ6GJC7sJUe4">colab</a>] 293 [<a href="https://www.youtube.com/watch?v=tbQaP07xQ4c&t=63s">video (~8min)</a>] 294 [<a href="files/publications/cvpr23_poster.pdf">poster</a>] 295 </p> 296 </article> 297 </div> 298 <div class="row"> 299 <article class="3u 12u$(xsmall) work-item"> 300 <a><img src="images/publications/testra.png" class="image fit thumb"></a> 301 </article> 302 <article class="9u 12u$(xsmall)"> 303 <h3>Real-time Online Video Detection with Temporal Smoothing Transformers<br /></h3> 304 <p> 305 <U>Yue Zhao</U>, Philipp Krähenbühl<br /> 306 European Conference on Computer Vision (ECCV), 2022 <br /> 307 [<a href="https://arxiv.org/pdf/2209.09236.pdf">arXiv</a>] 308 [<a href="https://github.com/zhaoyue-zephyrus/TeSTra">code</a>] 309 [<a href="files/publications/testra_poster.pdf">poster</a>] 310 </p> 311 </article> 312 </div> 313 <div class="row"> 314 <article class="3u 12u$(xsmall) work-item"> 315 <a><img src="images/publications/posec3d.png" class="image fit thumb"></a> 316 </article> 317 <article class="9u 12u$(xsmall)"> 318 <h3>Revisiting Skeleton-based Action Recognition<br /></h3> 319 <p> 320 Haodong Duan, <U>Yue Zhao</U>, Kai Chen, Dahua Lin, Bo Dai<br /> 321 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2022 <b style='color:orange'>(Oral, top-4.2%)</b><br /> 322 [<a href="https://arxiv.org/pdf/2104.13586.pdf">arXiv</a>] 323 [<a href="https://github.com/kennymckormick/pyskl">code</a>] 324 </p> 325 </article> 326 </div> 327 <!-- <div class="row"> 328 <article class="3u 12u$(xsmall) work-item"> 329 <a><img src="images/publications/omnisource.png" class="image fit thumb"></a> 330 </article> 331 <article class="9u 12u$(xsmall)"> 332 <h3>Omni-sourced Webly-supervised Learning for Video Recognition<br /></h3> 333 <p> 334 Haodong Duan, <U>Yue Zhao</U>, Yuanjun Xiong, Wentao Liu, Dahua Lin<br /> 335 European Conference on Computer Vision (ECCV), 2020 <br /> 336 <b style='color:orange'>Best single model on Kinetics-400 val (83.6%)</b>.<br /> 337 [<a href="https://arxiv.org/pdf/2003.13042.pdf">arXiv</a>] 338 [<a href="https://github.com/open-mmlab/mmaction/blob/master/MODEL_ZOO.md#omnisource">model</a>] 339 [<a href="https://venturebeat.com/2020/04/02/amazon-sensetime-omnisource-framework-web-data-video-recognition/">media</a>] 340 </p> 341 </article> 342 </div> --> 343 <div class="row"> 344 <article class="3u 12u$(xsmall) work-item"> 345 <a><img src="images/publications/finegym.png" class="image fit thumb"></a> 346 </article> 347 <article class="9u 12u$(xsmall)"> 348 <h3>FineGym: A Hierarchical Video Dataset for Fine-grained Action Understanding<br /></h3> 349 <p> 350 Dian Shao, <U>Yue Zhao</U>, Bo Dai, Dahua Lin<br /> 351 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020 <b style='color:orange'>
351(Oral, top-5.0%)</b>.<br /> 352 [<a href="https://arxiv.org/abs/2004.06704">arXiv</a>][<a href="https://sdolivia.github.io/FineGym/">project page</a>] 353 </p> 354 </article> 355 </div> 356 <!-- <div class="row"> 357 <article class="3u 12u$(xsmall) work-item"> 358 <a><img src="images/publications/tapos.png" class="image fit thumb"></a> 359 </article> 360 <article class="9u 12u$(xsmall)"> 361 <h3>Intra- and Inter-Action Understanding via Temporal Action Parsing<br /></h3> 362 <p> 363 Dian Shao, <U>Yue Zhao</U>, Bo Dai, Dahua Lin<br /> 364 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2020.<br /> 365 [<a href="https://arxiv.org/abs/2005.10229">arXiv</a>][<a href="https://sdolivia.github.io/TAPOS/">project page</a>] 366 </p> 367 </article> 368 </div> --> 369 <!-- <div class="row"> 370 <article class="3u 12u$(xsmall) work-item"> 371 <a><img src="images/publications/ssn.png" class="image fit thumb"></a> 372 </article> 373 <article class="9u 12u$(xsmall)"> 374 <h3>Temporal Action Detection with Structured Segment Networks</h3> 375 <p> 376 <U>Yue Zhao</U>, Yuanjun Xiong, Limin Wang, Zhirong Wu, Xiaoou Tang, Dahua Lin<br /> 377 International Journal on Computer Vision (IJCV), 2020.<br /> 378 [<a href="https://link.springer.com/article/10.1007/s11263-019-01211-2">Springer link</a>] 379 [<a href="http://openaccess.thecvf.com/content_ICCV_2017/papers/Zhao_Temporal_Action_Detection_ICCV_2017_paper.pdf">ICCV17 Version</a>] 380 [<a href="https://github.com/yjxiong/action-detection">code</a>] 381 [<a href="http://yjxiong.me/others/ssn/">project page</a>] 382 </p> 383 </article> 384 </div> --> 385 <!-- <div class="row"> 386 <article class="3u 12u$(xsmall) work-item"> 387 <a><img src="images/publications/trajectory_convolution.png" class="image fit thumb"></a> 388 </article> 389 <article class="9u 12u$(xsmall)"> 390 <h3>Trajectory Convolution for Action Recognition<br /></h3> 391 <p> 392 <U>Yue Zhao</U>, Yuanjun Xiong, Dahua Lin<br /> 393 Thirty-second Conference on Neural Information Processing Systems (NeurIPS), 2018.<br /> 394 [<a href="http://papers.nips.cc/paper/7489-trajectory-convolution-for-action-recognition">pdf</a>][<a href="https://github.com/open-mmlab/mmaction">code</a>] 395 </p> 396 </article> 397 </div> --> 398 <!-- <div class="row"> 399 <article class="3u 12u$(xsmall) work-item"> 400 <a><img src="images/publications/fifo.png" class="image fit thumb"></a> 401 </article> 402 <article class="9u 12u$(xsmall)"> 403 <h3>Find and Focus: Retrieve and Localize Video Events with Natural Language Queries<br /></h3> 404 <p> 405 Dian Shao*, Yu Xiong*, <U>Yue Zhao</U>, Qingqiu Huang, Yu Qiao, Dahua Lin<br /> 406 European Conference on Computer Vision (ECCV), 2018.<br /> 407 [<a href="http://openaccess.thecvf.com/content_ECCV_2018/papers/Dian_SHAO_Find_and_Focus_ECCV_2018_paper.pdf">pdf</a>] 408 [<a href="http://www.xiongyuxy.cn/projects/find_and_focus/">project page</a>] 409 </p> 410 </article> 411 </div> --> 412 <!-- <div class="row"> 413 <article class="3u 12u$(xsmall) work-item"> 414 <a><img src="images/publications/disentangle.png" class="image fit thumb"></a> 415 </article> 416 <article class="9u 12u$(xsmall)"> 417 <h3>Recognizing Actions by Disentangling Components of Dynamics<br /></h3> 418 <p> 419 <U>Yue Zhao</U>, Yuanjun Xiong, Dahua Lin<br /> 420 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018.<br /> 421 [<a href="http://openaccess.thecvf.com/content_cvpr_2018/papers/Zhao_Recognize_Actions_by_CVPR_2018_paper.pdf">pdf</a>] 422 [<a href="http://openaccess.thecvf.com/content_cvpr_2018/Supplemental/1067-supp.pdf">supplementary</a>] 423 </p> 424 </article> 425 </div> --> 426 <div class="row"> 427 <article class="3u 12u$(xsmall) work-item"> 428 <a><img src="images/publications/ssn.png" class="image fit thumb"></a> 429 </article> 430 <article class="9u 12u$(xsmall)"> 431 <h3>Temporal Action Detection with Structured Segment Networks</h3> 432 <p> 433 <U>Yue Zhao</U>, Yuanjun Xiong, Limin Wang, Zhirong Wu, Xiaoou Tang, Dahua Lin<br /> 434 International Conference on Computer Vision (ICCV), 2017.<br /> 435 [<a href="http://openaccess.thecvf.com/content_ICCV_2017/papers/Zhao_Temporal_Action_Detection_ICCV_2017_paper.pdf">pdf</a>] 436 [<a href="https://arxiv.org/pdf/1704.06228.pdf">arXiv</a>] 437 [<a href="https://link.springer.com/article/10.1007/s11263-019-01211-2">IJCV version</a>] 438 [<a href="https://github.com/yjxiong/action-detection">
438code</a>] 439 [<a href="http://yjxiong.me/others/ssn/">project page</a>] 440 </p> 441 </article> 442 </div> 443 <!-- <div class="row"> 444 <article class="3u 12u$(xsmall) work-item"> 445 <img src="images/publications/icassp.png" class="image fit thumb"></a> 446 </article> 447 <article class="9u 12u$(xsmall)"> 448 <h3>Recurrent Convolutional Neural Network for Speech Processing</h3> 449 <p> 450 <U>Yue Zhao</U>, Xingyu Jin, Xiaolin Hu<br /> 451 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2017.<br /> 452 [<a href="files/publications/icassp17.pdf">pdf</a>] 453 [<a href="https://github.com/zhaoyue-zephyrus/RecurrentConvNet-for-Speech">code</a>] 454 [<a href="files/publications/icassp17_poster.pdf">poster</a>] 455 </p> 456 </article> 457 </div> 458 <div class="row"> 459 <article class="3u 12u$(xsmall) work-item"> 460 <img src="images/publications/cinc.png" class="image fit thumb"></a> 461 </article> 462 <article class="9u 12u$(xsmall)"> 463 <h3>Heart Rate Estimation in Photoplethysmogram Signals using Nonlinear Model-Based Preprocessing</h3> 464 <p> 465 Federico Wadehn, <U>Yue Zhao</U>, Hans-Andrea Loeliger<br /> 466 IEEE Computing in Cardiology (CinC), 2015.<br /> 467 [<a href="http://www.cinc.org/archives/2015/pdf/0633.pdf">pdf</a>]</p> 468 </article> 469 </div> --> 470 </section> 471 472 <!-- Five --> 473 <section id="five"> 474 <h2>Education Experience<br /></h2> 475 <div class="row"> 476 <article class="2u 12u$(xsmall) work-item"> 477 <a><img src="images/education/utexas.png" class="image fit thumb"></a> 478 </article> 479 <article class="10u 12u$(xsmall)"> 480 <h4>The University of Texas at Austin, TX, USA</h4> 481 <p>August 2020 - August 2025<br /> 482 Ph.D. in Computer Science.</p> 483 </article> 484 </div> 485 <div class="row"> 486 <article class="2u 12u$(xsmall) work-item"> 487 <a><img src="images/professional/cuhk.png" class="image fit thumb"></a> 488 </article> 489 <article class="10u 12u$(xsmall)"> 490 <h4>The Chinese University of Hong Kong, HK SAR, China</h4> 491 <p>August 2017 - July 2020 <br /> 492 M.Phil. in Information Engineering.</p> 493 </article> 494 </div> 495 <div class="row"> 496 <article class="2u 12u$(xsmall) work-item"> 497 <a><img src="images/education/Technion.png" class="image fit thumb"></a> 498 </article> 499 <article class="10u 12u$(xsmall)"> 500 <h4>Israel Institute of Technology (Technion), Haifa, Israel</h4> 501 <p>July 2016 - August 2016<br /> 502 Visiting Student of <a href="http://int.technion.ac.il/academic-programs/summer-programs/summer-school-of-engineering-science/">Summer School of Engineering and Science</a>, fully funded by Technion. </p> 503 </article> 504 </div> 505 <div class="row"> 506 <article class="2u 12u$(xsmall) work-item"> 507 <a><img src="images/education/Tsinghua.png" class="image fit thumb"></a> 508 </article> 509 <article class="10u 12u$(xsmall)"> 510 <h4>Department of Electronic Engineering, Tsinghua University, Beijing, China</h4> 511 <p>August 2012 - July 2016<br /> 512 Bachelor of Engineering, magna cum laude.</p> 513 </article> 514 </div> 515 <div class="row"> 516 <article class="2u 12u$(xsmall) work-item"> 517 <a><img src="images/education/Tsinghua.png" class="image fit thumb"></a> 518 </article> 519 <article class="10u 12u$(xsmall)"> 520 <h4>School of Economics and Management, Tsinghua University, Beijing, China</h4> 521 <p>August 2013 - July 2016<br /> 522 Bachelor of Science (Second Degree) in Economics.</p> 523 </article> 524 </div> 525 <div class="row"> 526 <article class="2u 12u$(xsmall) work-item"> 527 <img src="images/education/eth_zurich.png" class="image fit thumb"> 528 </article> 529 <article class="10u 12u$(xsmall)"> 530 <h4>Department of Information Technology and Electrical Engineering (D-ITET), Swiss Federal Institute of Technology(ETH), Zürich, Switzerland </h4> 531 <p>September 2014 - Feburary 2015<br /> 532 Mobility student fully funded by China Scholarship Council (CSC).</p> 533 </article> 534 </div> 535 </section> 536 537 <!-- Six --> 538 <section id="six"> 539 <h2>Professional Experience<br /></h2> 540 <div class="row"> 541 <article class="2u 12u$(xsmall) work-item"> 542 <a><img src="images/professional/nvidia.png" class="image fit thumb"></a> 543 </article> 544 <article class="10u 12u$(xsmall)"> 545 <h4>NVIDIA Research, Santa Clara, CA, USA</h4> 546 <p>May 2024 - August 2024<br />
547 Research Scientist Intern</p> 548 </article> 549 </div> 550 <div class="row"> 551 <article class="2u 12u$(xsmall) work-item"> 552 <a><img src="images/professional/google.png" class="image fit thumb"></a> 553 </article> 554 <article class="10u 12u$(xsmall)"> 555 <h4>Google Research, Venice, CA, USA</h4> 556 <p>May 2023 - August 2023<br /> 557 Student Researcher</p> 558 </article> 559 </div> 560 <div class="row"> 561 <article class="2u 12u$(xsmall) work-item"> 562 <a><img src="images/professional/meta_ai.png" class="image fit thumb"></a> 563 </article> 564 <article class="10u 12u$(xsmall)"> 565 <h4>FAIR Labs, New York, NY, USA</h4> 566 <p>May 2022 - August 2022<br /> 567 Research Scientist Intern</p> 568 </article> 569 </div> 570 <div class="row"> 571 <article class="2u 12u$(xsmall) work-item"> 572 <a><img src="images/professional/aws.png" class="image fit thumb"></a> 573 </article> 574 <article class="10u 12u$(xsmall)"> 575 <h4>Amazon Web Services, Seattle, WA, USA</h4> 576 <p>June 2021 - August 2021<br /> 577 Applied Scientist Intern</p> 578 </article> 579 </div> 580 <div class="row"> 581 <article class="2u 12u$(xsmall) work-item"> 582 <a><img src="images/professional/cuhk.png" class="image fit thumb"></a> 583 </article> 584 <article class="10u 12u$(xsmall)"> 585 <h4>Multimedia Laboratory, The Chinese University of Hong Kong, HK SAR, China</h4> 586 <p>September 2016 - August 2017, July 2015 - September 2015<br /> 587 Junior Research Assistant</p> 588 </article> 589 </div> 590 </section> 591 592 <!-- Seven --> 593 <section id="Seven"> 594 <h2>Other projects<br /></h2> 595 <div class="row"> 596 <article class="3u 12u$(xsmall) work-item"> 597 <img src="images/projects/people_counting.png" class="image fit thumb"></a> 598 </article> 599 <article class="9u 12u$(xsmall)"> 600 <h3>People tracking using RGB-D videos.</h3> 601 <p> An undergraduate-level class project on people detection and tracking from RGB-D data collected by Kinect. 602 [<a href="https://www.youtube.com/watch?v=x8uYGiIhVPc">demo</a>] 603 </p> 604 </article> 605 </div> 606 </section> 607 608 <!-- Eight --> 609 <section id="eight"> 610 <h2>Miscellaneous<br /></h2> 611 </header> 612 <a id="pronouce">* </a>For non-Chinese speakers, the pronuciation for Zhao Yue (family name coming first is preferred) is close to ['dÊau 'ju:-eh]. 613 <p></p> 614 </section> 615 616 617 </div> 618 619 <!-- Footer --> 620 <footer id="footer"> 621 <div class="inner"> 622 <ul class="icons"> 623 <li><a href="https://github.com/zhaoyue-zephyrus" class="icon fa-brands fa-github"><span class="label">Github</span></a></li> 624 <li><a href="https://ai.stanford.edu/~yzz/blog/articles/npq.html" class="icon fa-solid fa-blog"><span class="label">Blog</span></a></li> 625 <li><a href="https://x.com/__yuezhao__" class="icon fa-brands fa-x-twitter"><span class="label">X</span></a></li> 626 <li><a href="https://scholar.google.com.hk/citations?user=6_U35tAAAAAJ&hl=en" class="icon fa-brands fa-google-scholar"><span class="label">GScholar</span></a></li> 627 <li><a href="https://www.linkedin.com/in/yue-zhao-450044b8" class="icon fa-brands fa-linkedin"><span class="label">LinkedIn</span></a></li> 628 <li><a href="mailto:[email protected]=Mail from My Homepage" class="icon fa-regular fa-envelope"><span class="label">Email</span></a></li> 629 <li><a href="https://orcid.org/0000-0003-2753-5921" class="icon fa-brands fa-orcid"><span class="label">ORCID</span></a></li> 630 </ul> 631 <ul class="copyright"> 632 <li>© Yue Zhao</li><li>Design: <a href="http://html5up.net">HTML5 UP</a></li> 633 <li>Background picture: Champs-Ãlysées, 2015.</li> 634 </ul> 635 </div> 636 </footer> 637 638 <!-- Scripts --> 639
639<script src="assets/js/jquery.min.js"></script>
639 640
640<script src="assets/js/jquery.poptrox.min.js"></script>
640 641
641<script src="assets/js/skel.min.js"></script>
641 642
642<script src="assets/js/util.js"></script>
642 643 <!--[if lte IE 8]>
643<script src="assets/js/ie/respond.min.js"></script>
643<![endif]--> 644
644<script src="assets/js/main.js"></script>
644 645 646 </body> 647</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.