1<!DOCTYPE html> 2<html lang="en"> 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=1.0"> 7 <meta name="google-site-verification" content="VqP1lG6BYTJ6SaMfg29j5rjWDMxLReOUu2HD_JeiSJA" /> 8 <title>Yaohui Wang</title> 9 10 <link rel='icon' href='favicon.ico' type='image/x-icon'/> 11 <link href="./css/bootstrap.min.css" rel="stylesheet"> 12 <link rel="stylesheet" href="./assets/academicons-1.7.0/css/academicons.css"/> 13 <link rel="stylesheet" href="./assets/font-awesome-4.7.0/css/font-awesome.min.css"/> 14 </head> 15 16 <body> 17 18 <div style="height:10px;"></div> 19 <!-- Navigation bar --> 20 21 <div class="navbar navbar-default navbar-fixed-top bg-info"> 22 <div class="container"> 23 <div class="navbar-header"> 24 25 <button class="navbar-toggle" type="button" data-toggle="collapse" data-target="#navbar-main"> 26 <span class="icon-bar"></span> 27 <span class="icon-bar"></span> 28 <span class="icon-bar"></span> 29 </button> 30 </div> 31 <div class="navbar-collapse collapse" id="navbar-main"> 32 33 <ul class="nav navbar-nav navbar-left"> 34 <li ><a href="index.html#aboutme">Home</a></li> 35 <li ><a href="#research">Research</a></li> 36 <li ><a href="#teaching">Teaching</a></li> 37 <li ><a href="group.html">Group Seminar</a></li> 38 </ul> 39 </div> 40 </div> 41 </div> 42 43 <div style="height:80px;"></div> 44 45 46 <!-- end of navigation bar --> 47 48 <!-- CONTENTS --> 49 <div class="container"> 50 <!-- Aboutme --> 51 <!--<div id="aboutme"></div> --> 52 <div style="font-family: Times New Roman" class="row"> 53 <div class="col-xs-6 col-sm-4 col-md-2"> 54 <a class="thumbnail"> 55 <img src="./img/me2026.jpg" alt="Yaohui Wang" class="img-rounded"> 56 </a> 57 </div> 58 59 <div class="col-xs-10 col-sm-6 col-md-4"> 60 <h1 class="text-info">Yaohui Wang çèæ</h1> 61 <h4 class="text-info">Ph.D. Inria</h4> 62 <h5> 63 <a href="mailto:[email protected]" class="text-info" title="e-Mail"><i class="fa fa-envelope-square fa-2x"></i></a> 64 <a href="https://scholar.google.com/citations?user=R7LyAb4AAAAJ&hl=EN" class="text-info" title="Google Scholar"><i class="ai ai-google-scholar-square ai-2x"></i></a> 65 <a href="https://github.com/wyhsirius" class="text-info" title="GitHub"><i class="fa fa-github-square fa-2x"></i></a> 66 <a href="https://www.linkedin.com/in/yaohui-wang-586983112" class="text-info" title="LinkedIn"><i class="fa fa-linkedin-square fa-2x"></i></a> 67 <a href="https://twitter.com/yaohuiwang_yh" class="text-info" title="Twitter"><i class="fa fa-twitter-square fa-2x"></i></a> 68 </h5> 69 70 <img src="./img/inria.png" alt="inria" style="width:20%"> 71 </div> 72 </div> 73 74 75 <p style="font-family: Times New Roman" class="text-body", align="justify"> 76 I am Research Scientist at <a href='' class="text-info">Shanghai Artificial Intelligence Laboratory</a>, and adjunct PhD advisor at <a href='' class="text-info">Shanghai Jiao Tong University</a>. My research focuses on applying machine learning on video generation models to simulate real-world.<br> 77 </p> 78 79 <p style="font-family: Times New Roman" class="text-body", align="justify"> 80 I obtained my PhD degree from <a href='https://www.inria.fr/en/stars' class="text-info">STARS team</a> at <a href='https://www.inria.fr/fr/centre-inria-sophia-antipolis-mediterranee' class="text-info">Inria Sophia Antipolis</a> advised by <a href='http://antitza.com' class="text-info">Antitza Dantcheva</a> and <a href='http://www-sop.inria.fr/members/Francois.Bremond/' class="text-info">Francois Bremond</a>. Prior to that, I completed Master <a href='https://sites.google.com/chalearn.org/ai-master' class='text-info'>Artificial Intelligence </a> directed by <a href='https://guyon.chalearn.org/', class='text-info'>Isabelle Guyon</a> and <a href='https://www.lri.fr/~sebag/', class='text-info'>Michèle Sebag</a> from <a href='https://www.universite-paris-saclay.fr/en', class='text-info'>Université Paris-Saclay</a>. 81 </p> 82 83 <p style="font-family: Times New Roman" class="text-body", align="justify"> 84 My works on video generation include <a href='https://vchitect.github.io/LaVie-project/' class="text-info">LaVie</a>, <a href='https://github.com/Vchitect/Latte' class="text-info">Latte</a>, <a href='https://github.com/Vchitect/SEINE' class="text-info">SEINE</a>, <a href='https://github.com/guoyww/animatediff/' class="text-info">AnimateDiff</a> and <a href='https://github.com/wyhsirius/LIA' class="text-info">LIA</a>. 85 </p> 86 87 <p style="font-family: Times New Roman"> 88 <strong>
88I am always seeking self-motivated PhD students and research interns to join my team. Feel free to <a href="mailto:[email protected]" class="text-info">contact me</a> if you are interested.</strong> 89 </p> 90 91 <!-- News --> 92 <div class="row" id="news" style="padding-top:60px; margin-top:-60px;"> 93 <div class="col-md-12"> 94 <h2>News</h2> 95 </div> 96 </div> 97 98 <div style="font-family: Times New Roman" class="row"> 99 <div class="col-sm-1 col-md-1"> 100 <span class="label label-info">03 / 2026</span> 101 </div> 102 <div class="col-sm-11 col-md-11"> 103 1 paper <a href="" class="text-info">AI killed the video star</a> accepted to <a href="" class="text-info">IJCV</a>. 104 </div> 105 </div> 106 107 <div style="font-family: Times New Roman" class="row"> 108 <div class="col-sm-1 col-md-1"> 109 <span class="label label-info">03 / 2026</span> 110 </div> 111 <div class="col-sm-11 col-md-11"> 112 2 papers <a href="" class="text-info">VDOT</a> and <a href="" class="text-info">PRISM</a> accepted to <a href="" class="text-info">CVPR 2026</a>. 113 </div> 114 </div> 115 116 <div style="font-family: Times New Roman" class="row"> 117 <div class="col-sm-1 col-md-1"> 118 <span class="label label-info">02 / 2026</span> 119 </div> 120 <div class="col-sm-11 col-md-11"> 121 1 paper <a href="" class="text-info">MiraMo</a> accepted to <a href="" class="text-info">TPAMI</a>. 122 </div> 123 </div> 124 125 <div style="font-family: Times New Roman" class="row"> 126 <div class="col-sm-1 col-md-1"> 127 <span class="label label-info">01 / 2026</span> 128 </div> 129 <div class="col-sm-11 col-md-11"> 130 2 papers <a href="" class="text-info">CineTrans</a> and <a href="" class="text-info">InternSVG</a> accepted to <a href="" class="text-info">ICLR 2026</a>. 131 </div> 132 </div> 133 134 135 <div style="font-family: Times New Roman" class="row"> 136 <div class="col-sm-1 col-md-1"> 137 <span class="label label-info">05 / 2025</span> 138 </div> 139 <div class="col-sm-11 col-md-11"> 140 2 papers <a href="" class="text-info">Mulan</a> and <a href="" class="text-info">TimeStep Master</a> accepted to <a href="" class="text-info">ICML 2025</a>. 141 </div> 142 </div> 143 144 <div style="font-family: Times New Roman" class="row"> 145 <div class="col-sm-1 col-md-1"> 146 <span class="label label-info">03 / 2025</span> 147 </div> 148 <div class="col-sm-11 col-md-11"> 149 1 paper <a href="" class="text-info">Latte</a> accepted to <a href="" class="text-info">TMLR 2025</a>. 150 </div> 151 </div> 152 153 <div style="font-family: Times New Roman" class="row"> 154 <div class="col-sm-1 col-md-1"> 155 <span class="label label-info">02 / 2025</span> 156 </div> 157 <div class="col-sm-11 col-md-11"> 158 3 papers <a href="" class="text-info">Cinemo</a>, <a href="" class="text-info">RAPO</a> and <a href="" class="text-info">Ouroboros3D</a> accepted to <a href="" class="text-info">CVPR 2025</a>. 159 </div> 160 </div> 161 162 163<!-- <div style="font-family: Times New Roman" class="row"> 164 <div class="col-sm-1 col-md-1"> 165 <span class="label label-info">11 / 2024</span> 166 </div> 167 <div class="col-sm-11 col-md-11"> 168 Invited Talk at <a href="https://univ-cotedazur.eu/events/sophia-summit" class="text-info">SophIA Summit 2024</a>. 169 </div> 170 </div> 171 172 <div style="font-family: Times New Roman" class="row"> 173 <div class="col-sm-1 col-md-1"> 174 <span class="label label-info">10 / 2024</span> 175 </div> 176 <div class="col-sm-11 col-md-11"> 177 1 paper <a href="https://vchitect.github.io/LaVie" class="text-info">LaVie</a> accepted to <a href="" class="text-info">IJCV</a>. 178 </div> 179 </div> 180 181 <div style="font-family: Times New Roman" class="row"> 182 <div class="col-sm-1 col-md-1"> 183 <span class="label label-info">10 / 2024</span> 184 </div> 185 <div class="col-sm-11 col-md-11"> 186 1 paper <a href="" class="text-info">4Diffusion</a> accepted to <a href="" class="text-info">NeurIPS 2024.</a> 187 </div> 188 </div> 189 190 <div style="font-family: Times New Roman" class="row"> 191 <div class="col-sm-1 col-md-1">
192 <span class="label label-info">08 / 2024</span> 193 </div> 194 <div class="col-sm-11 col-md-11"> 195 1 paper <a href="https://vchitect.github.io/LIA" class="text-info">LIA (Journal extension)</a> accepted to <a href="" class="text-info">TPAMI</a> and 1 paper <a href="https://vchitect.github.io/LEO-project" class="text-info">LEO</a> accepted to <a href="" class="text-info">IJCV</a>. 196 </div> 197 </div> 198 199 <div style="font-family: Times New Roman" class="row"> 200 <div class="col-sm-1 col-md-1"> 201 <span class="label label-info">02 / 2024</span> 202 </div> 203 <div class="col-sm-11 col-md-11"> 204 4 papers <a href="" class="text-info">VBench</a>, <a href="" class="text-info">EpiDiff</a>, <a href="" class="text-info">SinSR</a> and <a href="" class="text-info">Vlogger</a> accepted to <a href="" class="text-info">CVPR 2024</a>. 205 </div> 206 </div> 207 208 <div style="font-family: Times New Roman" class="row"> 209 <div class="col-sm-1 col-md-1"> 210 <span class="label label-info">01 / 2024</span> 211 </div> 212 <div class="col-sm-11 col-md-11"> 213 3 papers <a href="" class="text-info">SEINE</a>, <a href="" class="text-info">InternVid</a> and <a href="" class="text-info">AnimateDiff</a> accepted to <a href="https://aaai.org/aaai-conference/" class="text-info">ICLR 2024</a>. 214 </div> 215 </div> 216 217 <div style="font-family: Times New Roman" class="row"> 218 <div class="col-sm-1 col-md-1"> 219 <span class="label label-info">11 / 2023</span> 220 </div> 221 <div class="col-sm-11 col-md-11"> 222 2 papers <a href="" class="text-info">ConditionVideo</a> and <a href="" class="text-info">Diff-Text</a> accepted to <a href="https://aaai.org/aaai-conference/" class="text-info">AAAI 2024</a>. 223 </div> 224 </div> 225 <div style="height:3px;"></div> 226 227 <div style="font-family: Times New Roman" class="row"> 228 <div class="col-sm-1 col-md-1"> 229 <span class="label label-info">11 / 2023</span> 230 </div> 231 <div class="col-sm-11 col-md-11"> 232 Invited talk (Remote) at <a href='https://charlotteml.github.io' class="text-info">CharMLab, UNC Charlotte</a> on <a href='' class="text-info">Large-scale Video Generation Models</a>. 233 </div> 234 </div> 235 <div style="height:3px;"></div> 236 237 <div style="font-family: Times New Roman" class="row"> 238 <div class="col-sm-1 col-md-1"> 239 <span class="label label-info">07 / 2023</span> 240 </div> 241 <div class="col-sm-11 col-md-11"> 242 1 paper <a href="" class="text-info">LAC</a> accepted to <a href='' class="text-info">ICCV 2023</a>! 243 </div> 244 </div> 245 <div style="height:3px;"></div> 246 247 <div style="font-family: Times New Roman" class="row"> 248 <div class="col-sm-1 col-md-1"> 249 <span class="label label-info">04 / 2023</span> 250 </div> 251 <div class="col-sm-11 col-md-11"> 252 1 paper <a href="" class="text-info">Loris</a> accepted to <a href='https://icml.cc/Conferences/2023/Dates' class="text-info">ICML 2023</a>! 253 </div> 254 </div> 255 <div style="height:3px;"></div> 256 --> 257 <!-- 258 <div class="row"> 259 <div class="col-sm-1 col-md-1"> 260 <span class="label label-info">11 / 2022</span> 261 </div> 262 <div class="col-sm-11 col-md-11"> 263 One paper accepted to <a href='' class="text-info">TPAMI</a>! 264 </div> 265 </div> 266 <div style="height:3px;"></div> 267 268 <div class="row"> 269 <div class="col-sm-1 col-md-1"> 270 <span class="label label-info">11 / 2022</span> 271 </div> 272 <div class="col-sm-11 col-md-11"> 273 Our work <a href='' class="text-info">Latent Time Navigation (LTN)</a> has been accepted to <a href='' class="text-info">AAAI 2023</a>. 274 </div> 275 </div> 276 <div style="height:3px;"></div> 277 278 <div class="row"> 279 <div class="col-sm-1 col-md-1">
280 <span class="label label-info">08 / 2022</span> 281 </div> 282 <div class="col-sm-11 col-md-11"> 283 Invited <a href='https://www.bilibili.com/video/BV1bT411F7v4/?from=search' class="text-info">talk</a> at GAMES Seminar on <class="text-info">Image Animation</a>. 284 </div> 285 </div> 286 <div style="height:3px;"></div> 287 288 289 <div class="row"> 290 <div class="col-sm-1 col-md-1"> 291 <span class="label label-info">01 / 2022</span> 292 </div> 293 <div class="col-sm-11 col-md-11"> 294 Our work <a href='https://openreview.net/forum?id=7r6kDq0mK_' class="text-info">Latent Image Animator (LIA)</a> has been accepted to <a href='' class="text-info">ICLR 2022</a>, <a href='https://github.com/wyhsirius/LIA' class="text-info">code</a> released. 295 </div> 296 </div> 297 <div style="height:3px;"></div> 298 299 <div class="row"> 300 <div class="col-sm-1 col-md-1"> 301 <span class="label label-info">10 / 2021</span> 302 </div> 303 <div class="col-sm-11 col-md-11"> 304 I have successfully defended my PhD! 305 </div> 306 </div> 307 <div style="height:3px;"></div> 308 309 <div class="row"> 310 <div class="col-sm-1 col-md-1"> 311 <span class="label label-info">10 / 2021</span> 312 </div> 313 <div class="col-sm-11 col-md-11"> 314 Our work <a href='https://arxiv.org/pdf/2107.08580.pdf' class="text-info">UNIK</a> has been accepted to <a href='https://www.bmvc2021.com/' class="text-info">BMVC 2021</a> as oral presentation, <a href='https://github.com/YangDi666/UNIK' class="text-info">code</a> released. 315 </div> 316 </div> 317 <div style="height:3px;"></div> 318 319 <div class="row"> 320 <div class="col-sm-1 col-md-1"> 321 <span class="label label-info">03 / 2021</span> 322 </div> 323 <div class="col-sm-11 col-md-11"> 324 One paper accepted by <a href='http://cvpr2021.thecvf.com/' class="text-info">CVPR 2021</a>, <a href='https://github.com/chenhao2345/GCL' class="text-info">code</a> released. 325 </div> 326 </div> 327 <div class="row"> 328 <div class="col-sm-1 col-md-1"> 329 <span class="label label-info">01 / 2021</span> 330 </div> 331 <div class="col-sm-11 col-md-11"> 332 New work <a href='https://arXiv.org/pdf/2101.03049.pdf' class="text-info">InMoDeGAN</a> on Interpretable Unsupervised Video Generation is on arxiv. 333 </div> 334 </div> 335 <div style="height:3px;"></div> 336 337 338 <div class="row"> 339 <div class="col-sm-1 col-md-1"> 340 <span class="label label-info">12 / 2020</span> 341 </div> 342 <div class="col-sm-11 col-md-11"> 343 New work <a href='https://arxiv.org/pdf/2012.09071.pdf' class="text-info">GCL</a> on Unsupervised Person Re-id using GAN-based multi-view data augmentation is on arxiv. 344 </div> 345 </div> 346 <div style="height:3px;"></div> 347 348 <div class="row"> 349 <div class="col-sm-1 col-md-1"> 350 <span class="label label-info">12 / 2020</span> 351 </div> 352 <div class="col-sm-11 col-md-11"> 353 One paper <a href='https://arxiv.org/pdf/2012.09071.pdf' class="text-info">Selective Spatio-Temporal Aggregation Based Pose Refinement System</a> accepted by <a href='http://wacv2021.thecvf.com/home' class="text-info">WACV 2021</a>, <a href='https://github.com/YangDi666/SSTA-PRS' class="text-info">code</a> released. 354 </div> 355 </div> 356 <div style="height:3px;"></div> 357 358 <div class="row"> 359 <div class="col-sm-1 col-md-1"> 360 <span class="label label-info">02 / 2020</span> 361 </div> 362 <div class="col-sm-11 col-md-11"> 363 One paper <a href='' class="text-info">G³AN: Disentangling appearance and motion for video generation</a> accepted by <a href='http://cvpr2020.thecvf.com/' class="text-info">CVPR 2020</a>, <a href='https://github.com/wyhsirius/g3an-project' class="text-info">code</a> released. 364 </div> 365 </div> 366 <div style="height:3px;"></div> 367 368 <div class="row"> 369 <div class="col-sm-1 col-md-1">
370 <span class="label label-info">10 / 2019</span> 371 </div> 372 <div class="col-sm-11 col-md-11"> 373 I participated in PRAIRIE AI Summer School (<a href='https://project.inria.fr/paiss/speakers-2019/' class="text-info">PAISS 2019</a>) in Paris. 374 </div> 375 </div> 376 <div style="height:3px;"></div> 377 378 <div class="row"> 379 <div class="col-sm-1 col-md-1"> 380 <span class="label label-info">08 / 2019</span> 381 </div> 382 <div class="col-sm-11 col-md-11"> 383 Our paper <a href='' class="text-info">ImaGINator</a> has been accepted in <a href="http://wacv20.wacv.net" class="text-info">WACV 2020</a>. 384 </div> 385 </div> 386 <div style="height:3px;"></div> 387 388 <div class="row"> 389 <div class="col-sm-1 col-md-1"> 390 <span class="label label-success">02 / 2019</span> 391 </div> 392 <div class="col-sm-11 col-md-11"> 393 <a href=""></a> 394 </div> 395 </div> 396 <div style="height:3px;"></div> 397 398 <div class="row"> 399 <div class="col-sm-1 col-md-1"> 400 <span class="label label-info">09 / 2018</span> 401 </div> 402 <div class="col-sm-11 col-md-11"> 403 <a href="">Code</a> for <a href=""></a> is released. 404 </div> 405 </div> 406 <div style="height:3px;"></div> 407 408 <div class="row"> 409 <div class="col-sm-1 col-md-1"> 410 <span class="label label-info">08 / 2018</span> 411 </div> 412 <div class="col-sm-11 col-md-11"> 413 I visited Visual Information Processing and Learning group (<a href='http://vipl.ict.ac.cn' class="text-info">VIPL</a>) of Chinese Academy of Science in Beijing, China. 414 </div> 415 </div> 416 <div style="height:3px;"></div> 417 418 <div class="row"> 419 <div class="col-sm-1 col-md-1"> 420 <span class="label label-info">07 / 2018</span> 421 </div> 422 <div class="col-sm-11 col-md-11"> 423 I participated in PRAIRIE AI Summer School (<a href='https://project.inria.fr/paiss/speakers-2018/' class="text-info">PAISS 2018</a>) in Grenoble. 424 </div> 425 </div> 426 <div style="height:3px;"></div> 427 428 <div class="row"> 429 <div class="col-sm-1 col-md-1"> 430 <span class="label label-info">05 / 2018</span> 431 </div> 432 <div class="col-sm-11 col-md-11"> 433 Two papers accepted in <a href='https://eccv2018.org' class="text-info">ECCV 2018</a> workshops. 434 </div> 435 </div> 436 <div style="height:3px;"></div> 437 438 <div class="row"> 439 <div class="col-sm-1 col-md-1"> 440 <span class="label label-info">12 / 2017</span> 441 </div> 442 <div class="col-sm-11 col-md-11"> 443 I joined <a href='' class="text-info">STARS</a> team, Inria Sophia-antipolis as a PhD student. 444 </div> 445 </div> 446 <div style="height:3px;"></div> 447--> 448 <!-- end of news --> 449 450 <!-- Research --> 451 <div style="font-family: Times New Roman" class="row" id="research" style="padding-top:60px; margin-top:-60px;"> 452 <div class="col-md-12"> 453 <h2>Research</h2>(*equal contribution, <sup>†</sup>correspondance & project lead) 454 455<!--AI kill--> 456 <div class="row"> 457 <div class="col-xs-10 col-sm-4 col-md-4"> 458 <a class="thumbnail"> 459 <img src="img/audio-driven.png"> 460 </a> 461 </div> 462 <div class="col-xs-12 col-sm-8 col-md-8"> 463 <strong>AI killed the video star. Audio-driven diffusion model for expressive talking head generation</strong><br> 464 Baptiste Chopin, Tashvik Dhamija, Pranav Balaji, <u>Yaohui Wang</u><sup>†</sup>, Antitza Dantcheva<br> 465 <em>In IJCV 2026</em></br> 466 <a href="https://link.springer.com/article/10.1007/s11263-026-02766-7", class="text-info">Paper</a> | 467 <a href="https://arxiv.org/abs/2511.22488", class="text-info">Arxiv</a> | 468 <a href="", class='text-info'>Project page</a> | 469 <a href="", class="text-info"class="text-info">Code</a> | 470 471 <div style="height:30px;"></div> 472 </div> 473 </div> 474 475 476<!--LIAX--> 477 <div class="row"> 478 <div class="col-xs-10 col-sm-4 col-md-4"> 479 <a class="thumbnail"> 480 <img src="img/liax.gif"> 481 </a> 482 </div> 483 <div class="col-xs-12 col-sm-8 col-md-8"> 484 <strong>
484LIA-X: Interpretable Latent Portrait Animator</strong><br> 485 <u>Yaohui Wang</u>, Di Yang, Xinyuan Chen, Francois Bremond, Yu Qiao, Antitza Dantcheva<br> 486 <em>In arXiv 2025</em></br> 487 <a href="", class="text-info">Paper</a> | 488 <a href="https://arxiv.org/abs/2508.09959", class="text-info">Arxiv</a> | 489 <a href="https://wyhsirius.github.io/LIA-X-project/", class='text-info'>Project page</a> | 490 <a href="https://github.com/wyhsirius/LIA-X", class="text-info"class="text-info">Code</a> | 491 492 <div style="height:30px;"></div> 493 </div> 494 </div> 495 496<!--shotdirector--> 497 <div class="row"> 498 <div class="col-xs-10 col-sm-4 col-md-4"> 499 <a class="thumbnail"> 500 <img src="img/shotdirector.gif"> 501 </a> 502 </div> 503 <div class="col-xs-12 col-sm-8 col-md-8"> 504 <strong>ShotDirector: Directorially Controllable Multi-Shot Video Generation with Cinematographic Transitions</strong><br> 505 Xiaoxue Wu, Xinyuan Chen, <u>Yaohui Wang</u>, Yu Qiao<br> 506 <em>In CVPR 2026</em></br> 507 <a href="", class="text-info">Paper</a> | 508 <a href="https://arxiv.org/abs/2512.10286", class="text-info">Arxiv</a> | 509 <a href="https://uknowsth.github.io/ShotDirector/", class='text-info'>Project page</a> | 510 <a href="https://uknowsth.github.io/ShotDirector/", class="text-info"class="text-info">Code</a> | 511 512 <div style="height:30px;"></div> 513 </div> 514 </div> 515 516 517 518<!--vdot--> 519 <div class="row"> 520 <div class="col-xs-10 col-sm-4 col-md-4"> 521 <a class="thumbnail"> 522 <img src="img/vdot.gif"> 523 </a> 524 </div> 525 <div class="col-xs-12 col-sm-8 col-md-8"> 526 <strong>VDOT: Efficient Unified Video Creation via Optimal Transport Distillation</strong><br> 527 Yutong Wang, Haiyu Zhang, Tianfan Xue, Yu Qiao, <u>Yaohui Wang</u>, Chang Xu, Xinyuan Chen<br> 528 <em>In CVPR 2026</em></br> 529 <a href="", class="text-info">Paper</a> | 530 <a href="https://arxiv.org/abs/2512.06802", class="text-info">Arxiv</a> | 531 <a href="https://vdot-page.github.io/", class='text-info'>Project page</a> | 532 <a href="https://github.com/hhhh1138/VDOT", class="text-info"class="text-info">Code</a> | 533 534 <div style="height:30px;"></div> 535 </div> 536 </div> 537 538 539<!--vbench++--> 540 <div class="row"> 541 <div class="col-xs-10 col-sm-4 col-md-4"> 542 <a class="thumbnail"> 543 <img src="img/vbenchplus.png"> 544 </a> 545 </div> 546 <div class="col-xs-12 col-sm-8 col-md-8"> 547 <strong>Vbench++: Comprehensive and versatile benchmark suite for video generative models</strong><br> 548 Ziqi Huang, Fan Zhang, Xiaojie Xu, Yinan He, Jiashuo Yu, Ziyue Dong, Qianli Ma, Nattapol Chanpaisit, Chenyang Si, Yuming Jiang, <u>Yaohui Wang</u>, Xinyuan Chen, Ying-Cong Chen, Limin Wang, Dahua Lin, Yu Qiao, Ziwei Liu<br> 549 <em>In TPAMI 2025</em></br> 550 <a href="", class="text-info">Paper</a> | 551 <a href="https://arxiv.org/abs/2411.13503", class="text-info">Arxiv</a> | 552 <a href="https://vchitect.github.io/VBench-project/", class='text-info'>Project page</a> | 553 <a href="https://github.com/Vchitect/VBench", class="text-info"class="text-info">Code</a> | 554 555 <div style="height:30px;"></div> 556 </div> 557 </div> 558 559<!--CineTrans--> 560 <div class="row"> 561 <div class="col-xs-10 col-sm-4 col-md-4"> 562 <a class="thumbnail"> 563 <img src="img/cinetrans.gif"> 564 </a> 565 </div> 566 <div class="col-xs-12 col-sm-8 col-md-8"> 567 <strong>Cinetrans: Learning to generate videos with cinematic transitions via masked diffusion models</strong><br> 568 Xiaoxue Wu, Bingjie Gao, Yu Qiao, <u>Yaohui Wang</u>, Xinyuan Chen<br> 569 <em>In ICLR 2026</em></br> 570 <a href="", class="text-info">Paper</a> | 571 <a href="https://arxiv.org/abs/2508.11484", class="text-info">Arxiv</a> | 572 <a href="https://uknowsth.github.io/CineTrans/", class='text-info'>Project page</a> | 573 <a href="https://github.com/Vchitect/CineTrans", class="text-info"class="text-info">Code</a> | 574 575 <div style="height:30px;"></div> 576 </div> 577 </div> 578 579 580<!--RAPO++--> 581 <div class="row"> 582 <div class="col-xs-10 col-sm-4 col-md-4"> 583 <a class="thumbnail"> 584 <img src="img/rapoplus.gif"> 585 </a> 586 </div> 587 <div class="col-xs-12 col-sm-8 col-md-8"> 588 <strong>RAPO++: Cross-Stage Prompt Optimization for Text-to-Video Generation via Data Alignment and Test-Time Scaling</strong><br> 589 Bingjie Gao, Qianli Ma, Xiaoxue Wu, Shuai Yang, Guanzhou Lan, Haonan Zhao, Jiaxuan Chen, Qingyang Liu, Yu Qiao, Xinyuan Chen, <u>Yaohui Wang</u>, Li Niu<br> 590 <em>In arXiv 2025</em></br> 591 <a href="", class="text-info">Paper</a> | 592 <a href="https://arxiv.org/abs/2510.20206", class="text-info">Arxiv</a> | 593 <a href="https://whynothaha.github.io/RAPO_plus_github/", class='text-info'>Project page</a> | 594 <a href="https://github.com/Vchitect/RAPO", class="text-info"class="text-info">Code</a> | 595 596 <div style="height:30px;"></div> 597 </div> 598 </div> 599 600 601<!--GenHOI--> 602 <div class="row"> 603 <div class="col-xs-10 col-sm-4 col-md-4"> 604 <a class="thumbnail"> 605 <img src="img/genhoi.gif"> 606 </a> 607 </div> 608 <div class="col-xs-12 col-sm-8 col-md-8"> 609 <strong>GenHOI: Generalizing Text-driven 4D Human-Object Interaction Synthesis for Unseen Objects</strong><br> 610 Shujia Li, Haiyu Zhang, Xinyuan Chen, <u>Yaohui Wang</u><sup>†</sup>, Yutong Ban<sup>†</sup><br> 611 <em>In arXiv 2025</em></br> 612 <a href="", class="text-info">Paper</a> | 613 <a href="https://arxiv.org/abs/2506.15483", class="text-info">Arxiv</a> | 614 <a href="https://etach-qs.github.io/GenHOI_project/", class='text-info'>Project page</a> | 615 <a href="https://github.com/etach-qs/GenHOI", class="text-info"class="text-info">Code</a> | 616 617 <div style="height:30px;"></div> 618 </div> 619 </div> 620 621 622 623<!--AccVideo--> 624 <div class="row"> 625 <div class="col-xs-10 col-sm-4 col-md-4"> 626 <a class="thumbnail"> 627 <img src="img/accvideo.gif"> 628 </a> 629 </div> 630 <div class="col-xs-12 col-sm-8 col-md-8"> 631 <strong>AccVideo: Accelerating Video Diffusion Model with Synthetic Dataset</strong><br> 632 Haiyu Zhang, Xinyuan Chen, <u>Yaohui Wang</u>, Xihui Liu, Yunhong Wang, Yu Qiao<br> 633 <em>In arXiv 2025</em></br> 634 <a href="", class="text-info">Paper</a> | 635 <a href="https://arxiv.org/pdf/2503.19462", class="text-info">Arxiv</a> | 636 <a href="https://aejion.github.io/accvideo/", class='text-info'>Project page</a> | 637 <a href="https://github.com/aejion/AccVideo/", class="text-info"class="text-info">Code</a> | 638 <a href="https://huggingface.co/aejion/AccVideo", class="text-info"class="text-info">Hugging face</a> 639 <div style="height:30px;"></div> 640 </div> 641 </div> 642 643<!--Latte--> 644 <div class="row"> 645 <div class="col-xs-10 col-sm-4 col-md-4"> 646 <a class="thumbnail"> 647 <img src="img/latte.gif"> 648 </a> 649 </div> 650 <div class="col-xs-12 col-sm-8 col-md-8"> 651 <strong>Latte: Latent Diffusion Transformer for Video Generation</strong><br> 652 Xin Ma, <u>Yaohui Wang</u><sup>†</sup>, Xinyuan Chen, Gengyun Jia, Ziwei Liu, Yuan-fang Li, Cunjian Chen, Yu Qiao<br> 653 <em>In TMLR 2025</em></br> 654 <a href="", class="text-info">Paper</a> | 655 <a href="https://arxiv.org/abs/2401.03048v1", class="text-info">Arxiv</a> | 656 <a href="https://maxin-cn.github.io/latte_project/", class='text-info'>Project page</a> | 657 <a href="https://github.com/Vchitect/Latte", class="text-info"class="text-info">Code</a> |
658 <a href="https://huggingface.co/spaces/maxin-cn/Latte-1", class="text-info"class="text-info">Hugging face</a> 659 <div style="height:30px;"></div> 660 </div> 661 </div> 662 663<!--RAPO--> 664 <div class="row"> 665 <div class="col-xs-10 col-sm-4 col-md-4"> 666 <a class="thumbnail"> 667 <img src="img/rapo.png"> 668 </a> 669 </div> 670 <div class="col-xs-12 col-sm-8 col-md-8"> 671 <strong>The Devil is in the Prompts: Retrieval-Augmented Prompt Optimization for Text-to-Video Generation</strong><br> 672 Bingjie Gao, Xinyu Gao, Xiaoxue Wu, Yujie Zhou, Yu Qiao<sup>†</sup>, Li Niu<sup>†</sup>, Xinyuan Chen<sup>†</sup>, <u>Yaohui Wang</u><sup>†</sup><br> 673 <em>In Proc. CVPR, Nashville, 2025</em></br> 674 <a href="", class="text-info">Paper</a> | 675 <a href="", class="text-info">Arxiv</a> | 676 <a href="https://whynothaha.github.io/Prompt_optimizer/RAPO.html", class='text-info'>Project page</a> | 677 <a href="", class="text-info"class="text-info">Code</a> | 678 <a href="", class="text-info"class="text-info">Hugging face</a> 679 <div style="height:30px;"></div> 680 </div> 681 </div> 682 683<!--Ouroboros3D--> 684 <div class="row"> 685 <div class="col-xs-10 col-sm-4 col-md-4"> 686 <a class="thumbnail"> 687 <img src="img/ourrobos.gif"> 688 </a> 689 </div> 690 <div class="col-xs-12 col-sm-8 col-md-8"> 691 <strong>Ouroboros3D: Image-to-3D Generation via 3D-aware Recursive Diffusion</strong><br> 692 Hao Wen, Zehuan Huang, <u>Yaohui Wang</u>, Xinyuan Chen, Lu Sheng<br> 693 <em>In Proc. CVPR, Nashville, 2025</em></br> 694 <a href="", class="text-info">Paper</a> | 695 <a href="https://arxiv.org/abs/2406.03184", class="text-info">Arxiv</a> | 696 <a href="https://costwen.github.io/Ouroboros3D/", class='text-info'>Project page</a> | 697 <a href="https://github.com/Costwen/Ouroboros3D", class="text-info"class="text-info">Code</a> | 698 <a href="", class="text-info"class="text-info">Hugging face</a> 699 <div style="height:30px;"></div> 700 </div> 701 </div> 702 703<!--LaVie--> 704 <div class="row"> 705 <div class="col-xs-10 col-sm-4 col-md-4"> 706 <a class="thumbnail"> 707 <img src="img/lavie.gif"> 708 </a> 709 </div> 710 <div class="col-xs-12 col-sm-8 col-md-8"> 711 <strong>LaVie: High-Quality Video Generation with Cascaded Latent Diffusion Models</strong><br> 712 <u>Yaohui Wang*</u>, Xinyuan Chen*, Xin Ma*, Shangchen Zhou, Ziqi Huang, Yi Wang, Ceyuan Yang, Yinan He, Jiashuo Yu, Peiqing Yang, Yuwei Guo, Tianxing Wu, Chenyang Si, Yuming Jiang, Cunjian Chen, Chen Change Loy, Bo Dai, Dahua Lin<sup>†</sup>, Yu Qiao<sup>†</sup>, Ziwei Liu<sup>†</sup><br> 713 <em>In IJCV 2024</em></br> 714 <a href="", class="text-info">Paper</a> | 715 <a href="https://arxiv.org/abs/2309.15103", class="text-info">Arxiv</a> | 716 <a href="https://vchitect.github.io/LaVie-project/", class='text-info'>Project page</a> | 717 <a href="https://github.com/Vchitect/LaVie", class="text-info"class="text-info">Code</a> | 718 <a href="https://huggingface.co/spaces/Vchitect/LaVie", class="text-info"class="text-info">Hugging face</a> 719 <div style="height:30px;"></div> 720 </div> 721 </div> 722 723<!--Cinemo--> 724 <div class="row"> 725 <div class="col-xs-10 col-sm-4 col-md-4"> 726 <a class="thumbnail"> 727 <img src="img/cinemo.gif"> 728 </a> 729 </div> 730 <div class="col-xs-12 col-sm-8 col-md-8"> 731 <strong>Cinemo: Consistent and Controllable Image Animation with Motion Diffusion Models</strong><br> 732 Xin Ma, <u>Yaohui Wang</u><sup>†</sup>, Gengyun Jia, Xinyuan Chen, Tien-Tsin Wong, Yuan-fang Li, Cunjian Chen<sup>†</sup><br> 733 <em>In Proc. CVPR, Nashville, 2025</em></br> 734 <strong>Consistent and Controllable Image Animation with Motion Linear Diffusion Transformers</strong><br> 735 Xin Ma, <u>Yaohui Wang</u><sup>†</sup>, Gengyun Jia, Xinyuan Chen, Tien-Tsin Wong, Cunjian Chen<sup>†</sup><br> 736 <em>In TPAMI, 2026</em></br> 737 <a href="", class="text-info">Paper</a> | 738 <a href="", class="text-info">Paper(TPAMI version)</a> | 739 <a href="https://arxiv.org/abs/2407.15642", class="text-info">Arxiv</a> | 740 <a href="https://maxin-cn.github.io/cinemo_project/", class='text-info'>Project page</a> | 741 <a href="https://github.com/maxin-cn/Cinemo", class="text-info"class="text-info">Code</a> |
742 <a href="https://huggingface.co/spaces/maxin-cn/Cinemo", class="text-info"class="text-info">Hugging face</a> 743 <div style="height:30px;"></div> 744 </div> 745 </div> 746 747 748<!--4Diffusion--> 749 <div class="row"> 750 <div class="col-xs-10 col-sm-4 col-md-4"> 751 <a class="thumbnail"> 752 <img src="img/4diffusion.gif"> 753 </a> 754 </div> 755 <div class="col-xs-12 col-sm-8 col-md-8"> 756 <strong>4Diffusion: Multi-view Video Diffusion Model for 4D Generation</strong><br> 757 Haiyu Zhang, Xinyuan Chen, <u>Yaohui Wang</u>, Xihui Liu, Yunhong Wang, Yu Qiao<sup>†</sup><br> 758 <em>In Proc. NeurIPS, Vancouver, 2024</em></br> 759 <a href="", class="text-info">Paper</a> | 760 <a href="https://arxiv.org/pdf/2405.20674", class="text-info">Arxiv</a> | 761 <a href="https://aejion.github.io/4diffusion/", class='text-info'>Project page</a> | 762 <a href="https://github.com/aejion/4Diffusion", class="text-info"class="text-info">Code</a> 763 <div style="height:30px;"></div> 764 </div> 765 </div> 766 767<!--SEINE--> 768 <div class="row"> 769 <div class="col-xs-10 col-sm-4 col-md-4"> 770 <a class="thumbnail"> 771 <img src="img/seine.gif"> 772 </a> 773 </div> 774 <div class="col-xs-12 col-sm-8 col-md-8"> 775 <strong>SEINE: Short-to-Long Video Diffusion Model for Generative Transition and Prediction</strong><br> 776 Xinyuan Chen*, <u>Yaohui Wang*</u>, Lingjun Zhang, Shaobin Zhuang, Xin Ma, Jiashuo Yu, Yali Wang, Dahua Lin<sup>†</sup>, Yu Qiao<sup>†</sup>, Ziwei Liu<sup>†</sup><br> 777 <em>In Proc. ICLR, Vienna, 2024</em></br> 778 <a href="", class="text-info">Paper</a> | 779 <a href="https://arxiv.org/abs/2310.20700", class="text-info">Arxiv</a> | 780 <a href="https://vchitect.github.io/SEINE-project/", class='text-info'>Project page</a> | 781 <a href="https://github.com/Vchitect/SEINE", class="text-info"class="text-info">Code</a> | 782 <a href=https://huggingface.co/spaces/Vchitect/SEINE", class="text-info"class="text-info">Hugging face</a> 783 <div style="height:30px;"></div> 784 </div> 785 </div> 786 787<!--VBench--> 788 <div class="row"> 789 <div class="col-xs-10 col-sm-4 col-md-4"> 790 <a class="thumbnail"> 791 <img src="img/vbench.png"> 792 </a> 793 </div> 794 <div class="col-xs-12 col-sm-8 col-md-8"> 795 <strong>VBench: Comprehensive Benchmark Suite for Video Generative Models</strong><br> 796 Ziqi Huang, Yinan He, Jiashuo Yu, Fan Zhang, Chenyang Si, Yuming Jiang, Yuanhan Zhang, Tianxing Wu, Qingyang Jin, Nattapol Chanpaisit, <u>Yaohui Wang</u>, Xinyuan Chen, Limin Wang, Dahua Lin, Yu Qiao and Ziwei Liu<br> 797 <em>In Proc. CVPR, Seattle, 2024</em></br> 798 <a href="", class="text-info">Paper</a> | 799 <a href="https://arxiv.org/pdf/2311.17982.pdf", class="text-info">Arxiv</a> | 800 <a href="https://vchitect.github.io/VBench-project/", class='text-info'>Project page</a> | 801 <a href="https://github.com/Vchitect/VBench", class="text-info"class="text-info">Code</a> 802 <div style="height:30px;"></div> 803 </div> 804 </div> 805 806<!--EpiDiff--> 807 <div class="row"> 808 <div class="col-xs-10 col-sm-4 col-md-4"> 809 <a class="thumbnail"> 810 <img src="img/epidiff.png"> 811 </a> 812 </div> 813 <div class="col-xs-12 col-sm-8 col-md-8"> 814 <strong>EpiDiff:Enhancing Multi-View Synthesis via Localized Epipolar-Constrained Diffusion</strong><br> 815 Zehuan Huang, Hao Wen, Junting Dong, <u>Yaohui Wang</u>, Yangguang Li, Xinyuan Chen, Yan-pei Cao, Ding Liang, Yu Qiao, Bo Dai and Lu Sheng<br> 816 <em>In Proc. CVPR, Seattle, 2024</em></br> 817 <a href="", class="text-info">Paper</a> | 818 <a href="https://arxiv.org/pdf/2312.06725.pdf", class="text-info">Arxiv</a> | 819 <a href="https://huanngzh.github.io/EpiDiff/", class='text-info'>Project page</a> | 820 <a href="", class="text-info"class="text-info">Code</a> 821 <div style="height:30px;"></div> 822 </div> 823 </div> 824 825<!--SinSR--> 826 <div class="row"> 827 <div class="col-xs-10 col-sm-4 col-md-4"> 828 <a class="thumbnail"> 829 <img src="img/sinsr.png"> 830 </a> 831 </div> 832 <div class="col-xs-12 col-sm-8 col-md-8"> 833 <strong>SinSR: Diffusion-Based Image Super-Resolution in a Single Step</strong><br> 834 Yufei Wang, Wenhan Yang, Xinyuan Chen, <u>Yaohui Wang</u>, Lanqing Guo, Lap-Pui Chau, Ziwei Liu, Yu Qiao, Alex C. Kot and Bihan Wen<br> 835 <em>In Proc. CVPR, Seattle, 2024</em></br> 836 <a href="", class="text-info">Paper</a> | 837 <a href="https://arxiv.org/pdf/2311.14760.pdf", class="text-info">Arxiv</a> | 838 <a href="https://github.com/wyf0912/SinSR/", class='text-info'>Project page</a> | 839 <a href="https://github.com/wyf0912/SinSR/", class="text-info"class="text-info">Code</a> 840 <div style="height:30px;"></div> 841 </div> 842 </div> 843 844 845<!--vlogger--> 846 <div class="row"> 847 <div class="col-xs-10 col-sm-4 col-md-4"> 848 <a class="thumbnail"> 849 <img src="img/vlogger.png"> 850 </a> 851 </div> 852 <div class="col-xs-12 col-sm-8 col-md-8"> 853 <strong>Vlogger: Make Your Dream A Vlog</strong><br> 854 Shaobin Zhuang, Kunchang Li, Xinyuan Chen, <u>Yaohui Wang</u>, Ziwei Liu, Yu Qiao, Yali Wang<br> 855 <em>In Proc. CVPR, Seattle, 2024</em></br> 856 <a href="", class="text-info">Paper</a> | 857 <a href="https://arxiv.org/pdf/2401.09414.pdf", class="text-info">Arxiv</a> | 858 <a href="https://Vlogger.github.io", class='text-info'>Project page</a> | 859 <a href="https://github.com/Vchitect/Vlogger", class="text-info"class="text-info">Code</a> 860 <div style="height:30px;"></div> 861 </div> 862 </div> 863 864 865<!--difftext--> 866 <div class="row"> 867 <div class="col-xs-10 col-sm-4 col-md-4"> 868 <a class="thumbnail"> 869 <img src="img/difftext.png"> 870 </a> 871 </div> 872 <div class="col-xs-12 col-sm-8 col-md-8"> 873 <strong>Brush Your Text: Synthesize Any Scene Text on Images via Diffusion Model</strong><br> 874 Lingjun Zhang, Xinyuan Chen, <u>Yaohui Wang</u>, Yue Lu, Yu Qiao<br> 875 <em>In Proc. AAAI, Vancouver, 2024</em></br> 876 <a href="", class="text-info">Paper</a> | 877 <a href="", class="text-info">Arxiv</a> | 878 <a href="", class='text-info'>Project page</a> | 879 <a href="", class="text-info"class="text-info">Code</a> 880 <div style="height:30px;"></div> 881 </div> 882 </div> 883 884<!--ConditionVideo--> 885 <div class="row"> 886 <div class="col-xs-10 col-sm-4 col-md-4"> 887 <a class="thumbnail"> 888 <img src="img/conditionvideo.gif"> 889 </a> 890 </div> 891 <div class="col-xs-12 col-sm-8 col-md-8"> 892 <strong>ConditionVideo: Training-Free Condition-Guided Text-to-Video Generation</strong><br> 893 Bo Peng, Xinyuan Chen, <u>Yaohui Wang</u>
893, Chaochao Lu, Yu Qiao<br> 894 <em>In Proc. AAAI, Vancouver, 2024</em></br> 895 <a href="", class="text-info">Paper</a> | 896 <a href="https://arxiv.org/abs/2310.07697", class="text-info">Arxiv</a> | 897 <a href="https://pengbo807.github.io/conditionvideo-website/", class='text-info'>Project page</a> | 898 <a href="https://github.com/pengbo807/ConditionVideo", class="text-info"class="text-info">Code</a> 899 <div style="height:30px;"></div> 900 </div> 901 </div> 902 903<!--LAC--> 904 <div class="row"> 905 <div class="col-xs-10 col-sm-4 col-md-4"> 906 <a class="thumbnail"> 907 <img src="img/lac.gif"> 908 </a> 909 </div> 910 <div class="col-xs-12 col-sm-8 col-md-8"> 911 <strong>LAC: Latent Action Composition for Skeleton-based Action Segmentation</strong><br> 912 Di Yang, <u>Yaohui Wang</u><sup>†</sup>, Antitza Dantcheva, Quan Kong, Lorenzo Garattoni, Gianpiero Francesca, Francois Bremond. <sup>†</sup>corresponding author<br> 913 <em>In Proc. ICCV, Paris, 2023</em></br> 914 <a href="https://openaccess.thecvf.com/content/ICCV2023/papers/Yang_LAC_-_Latent_Action_Composition_for_Skeleton-based_Action_Segmentation_ICCV_2023_paper.pdf", class="text-info">Paper</a> | 915 <a href="https://arxiv.org/abs/2308.14500", class="text-info">Arxiv</a> | 916 <a href="https://walker1126.github.io/LAC/", class='text-info'>Project page</a> | 917 <a href="https://github.com/walker1126/Latent_Action_Composition", class="text-info"class="text-info">Code</a> 918 <div style="height:30px;"></div> 919 </div> 920 </div> 921 922<!--InternVid--> 923 <div class="row"> 924 <div class="col-xs-10 col-sm-4 col-md-4"> 925 <a class="thumbnail"> 926 <img src="img/internvid.jpeg"> 927 </a> 928 </div> 929 <div class="col-xs-12 col-sm-8 col-md-8"> 930 <strong>InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation</strong><br> 931 Yi Wang*, Yinan He*, Yuzhuo Li*, Kunchang Li, Jiashuo Yu, Xin Ma, Xinyuan Chen, <u>Yaohui Wang</u>, Ping Luo, Ziwei Liu, Yali Wang<sup>†</sup>, Limin Wang<sup>†</sup>, Yu Qiao<sup>†</sup><br> 932 <em>In Proc. ICLR, Vienna, 2024</em></br> 933 <a href="", class="text-info">Paper</a> | 934 <a href="https://arxiv.org/pdf/2307.06942.pdf", class="text-info">Arxiv</a> | 935 <a href="https://github.com/OpenGVLab/InternVideo/tree/main/Data/InternVid", class='text-info'>Project page</a> | 936 <a href="https://github.com/OpenGVLab/InternVideo/tree/main/Data/InternVid", class="text-info"class="text-info">Code</a> 937 <div style="height:30px;"></div> 938 </div> 939 </div> 940 941<!--AnimateDiff--> 942 <div class="row"> 943 <div class="col-xs-10 col-sm-4 col-md-4"> 944 <a class="thumbnail"> 945 <img src="img/animatediff.gif"> 946 </a> 947 </div> 948 <div class="col-xs-12 col-sm-8 col-md-8"> 949 <strong>AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning</strong><br> 950 Yuwei Guo, Ceyuan Yang, Anyi Rao, <u>Yaohui Wang</u>, Yu Qiao, Dahua Lin, Bo Dai<br> 951 <em>In Proc. ICLR, Vienna, 2024</em></br> 952 <a href="", class="text-info">Paper</a> | 953 <a href="https://arxiv.org/abs/2307.04725", class="text-info">Arxiv</a> | 954 <a href="https://animatediff.github.io/", class='text-info'>Project page</a> | 955 <a href="https://github.com/guoyww/animatediff/", class="text-info"class="text-info">Code</a> 956 <div style="height:30px;"></div> 957 </div> 958 </div> 959 960<!--LEO--> 961 <div class="row"> 962 <div class="col-xs-10 col-sm-4 col-md-4"> 963 <a class="thumbnail"> 964 <img src="img/leo.gif"> 965 </a> 966 </div> 967 <div class="col-xs-12 col-sm-8 col-md-8"> 968 <strong>LEO: Generative Latent Image Animator for Human Video Synthesis</strong><br> 969 <u>Yaohui Wang</u>, Xin Ma, Xinyuan Chen, Cunjian Chen, Antitza Dancheva, Bo Dai, Yu Qiao<br> 970 <em>In IJCV 2024</em></br> 971 <a href="", class="text-info">Paper</a> | 972 <a href="https://arxiv.org/abs/2305.03989", class="text-info">Arxiv</a> | 973 <a href="https://wyhsirius.github.io/LEO-project/", class='text-info'>Project page</a> | 974 <a href="https://github.com/wyhsirius/LEO", class="text-info"class="text-info">Code</a> 975 <div style="height:30px;"></div> 976 </div> 977 </div> 978 979<!--HDAE--> 980 <div class="row"> 981 <div class="col-xs-10 col-sm-4 col-md-4"> 982 <a class="thumbnail"> 983 <img src="img/hdae.png"> 984 </a> 985 </div> 986 <div class="col-xs-12 col-sm-8 col-md-8"> 987 <strong>Hierarchical Diffusion Autoencoders and Disentangled Image Manipulation</strong><br> 988 Zeyu Lu, Chengyue Wu, Xinyuan Chen, <u>Yaohui Wang</u>, Lei Bai, Yu Qiao, Xihui Liu<br> 989 <em>In Proc. WACV, Hawaii, 2024</em></br> 990 <a href="", class="text-info">Paper</a> | 991 <a href="https://arxiv.org/abs/2304.11829", class="text-info">Arxiv</a> | 992 <a href="", class='text-info'>Project page</a> | 993 <a href="", class="text-info"class="text-info">Code</a> 994 <div style="height:30px;"></div> 995 </div> 996 </div> 997 998<!--LORIS--> 999 <div class="row"> 1000 <div class="col-xs-10 col-sm-4 col-md-4"> 1001 <a class="thumbnail"> 1002 <img src="img/loris.png"> 1003 </a> 1004 </div> 1005 <div class="col-xs-12 col-sm-8 col-md-8"> 1006 <strong>Long-Term Rhythmic Video Soundtracker</strong><br> 1007 Jiashuo Yu, <u>Yaohui Wang</u>, Xinyuan Chen, Xiao Sun, Yu Qiao<br> 1008 <em>In Proc. ICML, Hawaii, 2023</em></br> 1009 <a href="", class="text-info">Paper</a> | 1010 <a href="https://arxiv.org/abs/2305.01319", class="text-info">Arxiv</a> | 1011 <a href='', class='text-info'>Project page</a> | 1012 <a href="https://github.com/OpenGVLab/LORIS", class="text-info"class="text-info">Code</a> 1013 <div style="height:30px;"></div> 1014 </div> 1015 </div> 1016 1017 1018<!--LTN--> 1019 <div class="row"> 1020 <div class="col-xs-10 col-sm-4 col-md-4"> 1021 <a class="thumbnail"> 1022 <img src="img/LTN.png"> 1023 </a> 1024 </div> 1025 <div class="col-xs-12 col-sm-8 col-md-8"> 1026 <strong>Self-supervised Video Representation Learning via Latent Time Navigation</strong><br> 1027 Di Yang, <u>Yaohui Wang</u>, Quan Kong, Antitza Dantcheva, Lorenzo Garattoni, Gianpiero Francesca and François Brémond<br> 1028 <em>In Proc. AAAI, Washington, 2023</em></br> 1029 <a href="", class="text-info">Paper</a> | 1030 <a href="", class="text-info">Arxiv</a> | 1031 <a href='', class='text-info'>Project page</a> | 1032 <a href="", class="text-info"class="text-info">Code</a> 1033 <div style="height:30px;"></div> 1034 </div> 1035 </div> 1036 1037<!--ViA--> 1038 <div class="row"> 1039 <div class="col-xs-10 col-sm-4 col-md-4"> 1040 <a class="thumbnail"> 1041 <img src="img/via.gif"> 1042 </a> 1043 </div> 1044 <div class="col-xs-12 col-sm-8 col-md-8"> 1045 <strong>
1045ViA: View-invariant Skeleton Action Representation Learning via Motion Retargeting</strong><br> 1046 Di Yang, <u>Yaohui Wang</u><sup>†</sup>, Antitza Dantcheva, Lorenzo Garattoni, Gianpiero Francesca and François Brémond<br> 1047 <em>In IJCV, 2023</em></br> 1048 <a href="", class="text-info">Paper</a> | 1049 <a href="https://arxiv.org/abs/2209.00065", class="text-info">Arxiv</a> | 1050 <a href='https://walker-a11y.github.io/ViA-project/', class='text-info'>Project page</a> | 1051 <a href="", class="text-info"class="text-info">Code</a> 1052 <div style="height:30px;"></div> 1053 </div> 1054 </div> 1055 1056 1057<!--LIA--> 1058 <div class="row"> 1059 <div class="col-xs-10 col-sm-4 col-md-4"> 1060 <a class="thumbnail"> 1061 <img src="img/LIA.gif"> 1062 </a> 1063 </div> 1064 <div class="col-xs-12 col-sm-8 col-md-8"> 1065 <strong>Latent Image Animator: Learning to Animate Images via Latent Space Navigation</strong><br> 1066 <u>Yaohui Wang</u>, Di Yang, Francois Bremond and Antitza Dantcheva<br> 1067 <em>In Proc. ICLR, Virtual, 2022</em></br> 1068 <strong>LIA: Latent Image Animator</strong><br> 1069 <u>Yaohui Wang</u>, Di Yang, Francois Bremond and Antitza Dantcheva<br> 1070 <em>In TPAMI, 2024</em></br> 1071 <a href="https://openreview.net/pdf?id=7r6kDq0mK_", class="text-info">Paper</a> | 1072 <a href="https://ieeexplore.ieee.org/document/10645735/", class="text-info">Paper (TPAMI version)</a> | 1073 <a href="https://arxiv.org/abs/2203.09043", class="text-info">Arxiv</a> | 1074 <a href='http://wyhsirius.github.io/LIA-project/', class='text-info'>Project page</a> | 1075 <a href="https://github.com/wyhsirius/LIA", class="text-info"class="text-info">Code</a> 1076 <div style="height:30px;"></div> 1077 </div> 1078 </div> 1079 1080 1081<!--UNIK--> 1082 <div class="row"> 1083 <div class="col-xs-10 col-sm-4 col-md-4"> 1084 <a class="thumbnail"> 1085 <img src="img/unik.png"> 1086 </a> 1087 </div> 1088 <div class="col-xs-12 col-sm-8 col-md-8"> 1089 <strong>UNIK: A Unified Framework for Real-world Skeleton-based Action Recognition</strong><br> 1090 Di Yang*, <u>Yaohui Wang*</u>, Antitza Dantcheva, Lorenzo Garattoni, Gianpiero Francesca and Francois Bremond. *equal contribution<br> 1091 <em>In Proc. BMVC, Virtual, 2021 (Oral)</em></br> 1092 <a href="http://www-sop.inria.fr/members/Francois.Bremond/Postscript/Di-bmvc2021.pdf", class="text-info">Paper</a> | 1093 <a href="https://arxiv.org/pdf/2107.08580.pdf", class="text-info">Arxiv</a> | 1094 <a href='https://yangdi666.github.io/UNIK-project/', class='text-info'>Project page</a> | 1095 <a href="https://github.com/YangDi666/UNIK", class="text-info"class="text-info">Code</a> 1096 <div style="height:30px;"></div> 1097 </div> 1098 </div> 1099 1100<!-- inmodegan --> 1101 <div class="row"> 1102 <div class="col-xs-10 col-sm-4 col-md-4"> 1103 <a class="thumbnail"> 1104 <img src="img/inmodegan.gif"> 1105 </a> 1106 </div> 1107 <div class="col-xs-12 col-sm-8 col-md-8"> 1108 <strong>InMoDeGAN: Interpretable Motion Decomposition Generative Adversarial Network for Video Generation</strong><br> 1109 <u>Yaohui Wang</u>, Francois Bremond, and Antitza Dantcheva<br> 1110 <em>arXiv:2101.03049</em></br> 1111 <a href="https://arxiv.org/pdf/2101.03049.pdf", class="text-info">Arxiv</a> | 1112 <a href="https://wyhsirius.github.io/InMoDeGAN/", class="text-info"class="text-info">Project page</a> | 1113 <a href="https://github.com/wyhsirius/InMoDeGAN-project", class="text-info"class="text-info">Code</a> 1114 <div style="height:30px;"></div> 1115 </div> 1116 </div> 1117<!-- end of inmodegan --> 1118 1119 1120<!-- GCL --> 1121 <div class="row"> 1122 <div class="col-xs-10 col-sm-4 col-md-4"> 1123 <a class="thumbnail"> 1124 <img src="./img/gcl.png"> 1125 </a> 1126 </div> 1127 <div class="col-xs-12 col-sm-8 col-md-8"> 1128 <strong>Joint Generative and Contrastive Learning for Unsupervised Person Re-identification</strong><br> 1129 Hao Chen*, <u>Yaohui Wang*</u>, Benoit Lagadec, Antitza Dantcheva, and Francois Bremond. *equal contribution<br> 1130 <em>In Proc. CVPR, Virtual,</em> 2021.<br> 1131 <a href="https://openaccess.thecvf.com/content/CVPR2021/papers/Chen_Joint_Generative_and_Contrastive_Learning_for_Unsupervised_Person_Re-Identification_CVPR_2021_paper.pdf", class="text-info">Paper</a> | 1132 <a href="https://github.com/chenhao2345/GCL", class="text-info">Code</a><br> 1133 <strong>Learning Invariance from Generated Variance or Unsupervised Person Re-identification</strong><br> 1134 <em>In IEEE TPAMI, </em> 2022.<br> 1135 <a href="", class="text-info">Paper</a> | 1136 <a href="https://github.com/chenhao2345/GCL", class="text-info">Code</a> 1137 <div style="height:30px;"></div> 1138 </div> 1139 </div> 1140<!-- end of GCL --> 1141 1142 1143<!-- SSTA-PRS --> 1144 <div class="row"> 1145 <div class="col-xs-10 col-sm-4 col-md-4"> 1146 <a class="thumbnail"> 1147 <img src="./img/ssta.png"> 1148 </a> 1149 </div> 1150 <div class="col-xs-12 col-sm-8 col-md-8"> 1151 <strong>Selective Spatio-Temporal Aggregation Based Pose Refinement System</strong><br> 1152 Di Yang, Rui Dai, <u>Yaohui Wang</u>, Rupayan Mallick, Luca Minciullo, Gianpiero Francesca, and Francois Bremond<br> 1153 <em>In Proc. WACV, Virtual,</em> 2021.<br> 1154 <a href="https://arxiv.org/pdf/2011.05358.pdf", class="text-info">Paper</a> | 1155 <a href="https://github.com/YangDi666/SSTA-PRS", class="text-info">Code</b></a> 1156 <div style="height:30px;"></div> 1157 </div> 1158 </div> 1159<!-- end of GCL --> 1160 1161 1162<!-- G3AN --> 1163 <div class="row"> 1164 <div class="col-xs-10 col-sm-4 col-md-4"> 1165 <a class="thumbnail" href=""> 1166 <img src="./img/g3an.png" alt="G³AN: Disentangling appearance and motion for video generation"> 1167 </a> 1168 </div> 1169 <div class="col-xs-12 col-sm-8 col-md-8"> 1170 <strong>G³AN: Disentangling appearance and motion for video generation</strong><br> 1171 <u>Yaohui Wang</u>, Piotr Bilinski, Francois Bremond, and Antitza Dantcheva<br> 1172 <em>In Proc. CVPR, Seattle, US,</em> 2020.<br> 1173 <em>In LUV-CVPR Workshop, Seattle, US,</em> 2020. (Oral Presentation)<br> 1174 <a href="https://arxiv.org/pdf/1912.05523.pdf", class="text-info">Paper</button></a> | 1175 <a href="https://wyhsirius.github.io/G3AN/". class="text-info">Project page</a> | 1176 <a href="https://github.com/wyhsirius/g3an-project". class="text-info">Code</a> | 1177 <a href="https://www.youtube.com/watch?v=jEXbmxGKFfQ", class="text-info">Video</a>
1178 <div style="height:30px;"></div> 1179 </div> 1180 </div> 1181<!-- end of G3AN --> 1182 1183<!-- ImaGINator --> 1184 <div class="row"> 1185 <div class="col-xs-10 col-sm-4 col-md-4"> 1186 <a class="thumbnail" href=""> 1187 <img src="./img/imaginator.png" alt="ImaGINator: Conditional Spatio-Temporal GAN for Video Generation"> 1188 </a> 1189 </div> 1190 <div class="col-xs-12 col-sm-8 col-md-8"> 1191 <strong>ImaGINator: Conditional Spatio-Temporal GAN for Video Generation</strong><br> 1192 <u>Yaohui Wang</u>, Piotr Bilinski, Francois Bremond, and Antitza Dantcheva<br> 1193 <em>In Proc. WACV, Aspen, US,</em> 2020.<br> 1194 <a href="http://openaccess.thecvf.com/content_WACV_2020/papers/WANG_ImaGINator_Conditional_Spatio-Temporal_GAN_for_Video_Generation_WACV_2020_paper.pdf", class="text-info">Paper</a> | 1195 <a href="https://github.com/wyhsirius/ImaGINator", class="text-info">Code</a> | 1196 <a href="img/imaginator.mp4", class="text-info">Video</a> 1197 <div style="height:30px;"></div> 1198 </div> 1199 </div> 1200<!-- end of ImaGINator --> 1201 1202<!--FG20-->
1202 1203 <div class="row"> 1204 <div class="col-xs-10 col-sm-4 col-md-4"> 1205 <a class="thumbnail" href=""> 1206 <img src="./img/fg20.png" alt="A video is worth more than 1000 lies. Comparing 3DCNN approaches for detecting deepfakes"> 1207 </a> 1208 </div> 1209 <div class="col-xs-12 col-sm-8 col-md-8"> 1210 <strong>A video is worth more than 1000 lies. Comparing 3DCNN approaches for detecting deepfakes</strong><br> 1211 <u>Yaohui Wang</u> and Antitza Dantcheva<br> 1212 <em>In Proc. FG, Buenos Aires, Argentina,</em> 2020.<br> 1213 <a href="http://antitza.com/19.wang_fg_20.pdf", class="text-info">Paper</a> 1214 <div style="height:30px;"></div> 1215 </div> 1216 </div> 1217 1218<!--FA2F--> 1219 <div class="row"> 1220 <div class="col-xs-10 col-sm-4 col-md-4"> 1221 <a class="thumbnail" href=""> 1222 <img src="./img/fa2f.png" alt="From attribute-labels to faces: face generation using a conditional generative adversarial network"> 1223 </a> 1224 </div> 1225 <div class="col-xs-12 col-sm-8 col-md-8"> 1226 <strong>From attribute-labels to faces: face generation using a conditional generative adversarial network</strong><br> 1227 <u>Yaohui Wang</u>, Antitza Dantcheva and Francois Bremond<br> 1228 <em>In Proc. ECCV Workshop, Munich, Germany,</em> 2018.<br> 1229 <a href="https://openaccess.thecvf.com/content_ECCVW_2018/papers/11132/Wang_From_attribute-labels_to_faces_face_generation_using_a_conditional_generative_ECCVW_2018_paper.pdf", class="text-info">Paper</a> 1230 <div style="height:30px;"></div> 1231 </div> 1232 </div> 1233 1234<!--eccvw18--> 1235 <div class="row"> 1236 <div class="col-xs-10 col-sm-4 col-md-4"> 1237 <a class="thumbnail" href=""> 1238 <img src="./img/eccvw18.png" alt="Comparing methods for assessment of facial dynamics in patients with major neurocognitive disorders"> 1239 </a> 1240 </div> 1241 <div class="col-xs-12 col-sm-8 col-md-8"> 1242 <strong>Comparing methods for assessment of facial dynamics in patients with major neurocognitive disorders</strong><br> 1243 <u>Yaohui Wang</u>, Antitza Dantcheva, Francois Bremond and Piotr Bilinski<br> 1244 <em>In Proc. ECCV Workshop, Munich, Germany,</em> 2018.<br> 1245 <a href="https://openaccess.thecvf.com/content_ECCVW_2018/papers/11134/Wang_Comparing_methods_for_assessment_of_facial_dynamics_in_patients_with_ECCVW_2018_paper.pdf", class="text-info">Paper</a> 1246 <div style="height:30px;"></div> 1247 </div> 1248 </div> 1249 1250 1251 <div class="row" id="teaching" style="padding-top:60px; margin-top:-60px;"> 1252 <div class="col-md-12"> 1253 <h2>Professional activities</h2> 1254 <div class="row"> 1255 <div class="col-sm-1 col-md-1"> 1256 <span class="label label-info"> Reviewer </span> 1257 </div> 1258 <div class="col-sm-11 col-md-11"> 1259 SIGGRAPH 2022, CVPR 2022/2021, ECCV 2022/2020, WACV 2020 ... 1260 </div> 1261 </div> 1262 1263 </div> 1264 </div> 1265 1266 1267<div class="row" id="teaching" style="padding-top:60px; margin-top:-60px;"> 1268 <div class="col-md-12"> 1269 <h2>PhD Thesis</h2> 1270 <div class="col-xs-6 col-sm-4 col-md-3"> 1271 <a class="thumbnail"> 1272 <img src="./img/thesis.png" alt="thesis" class="img-rounded"> 1273 </a> 1274 1275 </div> 1276 <div class="col-xs-12 col-sm-8 col-md-8"> 1277 <strong>Learning to Generate Human Videos</strong><br> 1278 Yaohui Wang<br> 1279 <a href="https://hal.inria.fr/tel-03551913v1", class="text-info">Thesis</a><br><br> 1280 Defense Jury: 1281 <ul> 1282 <li><a href="http://www-sop.inria.fr/members/George.Drettakis/", class="text-info">George Drettakis (President)</a>, Inria Sophia Antipolis, France</li> 1283 <li><a href="https://www.di.ens.fr/~laptev/", class="text-info">Ivan Laptev</a>, Inria Paris/Ãcole normale supérieure, France </li> 1284 <li><a href="http://elisaricci.eu/", class="text-info">Elisa Ricci</a>, University of Trento, Italy</li> 1285 <li><a href="https://people.ucas.edu.cn/~sgshan?language=en", class="text-info">Shiguang Shan</a>, Chinese Academy of Sciences, China</li> 1286 <li><a href="http://www.stulyakov.com/", class="text-info">Sergey Tulyakov</a>, Snap Research, USA</li> 1287 <li><a href="https://scholar.google.com/citations?user=RKt2sFIAAAAJ&hl=zh-CN", class="text-info">Panayiotis Georgiou</a>, Apple Inc./University of Southern California, USA</li> 1288 </ul> 1289 1290 1291 </div> 1292 </div> 1293 1294 1295 1296 <!-- Teaching --> 1297 <div class="row" id="teaching" style="padding-top:60px; margin-top:-60px;"> 1298 <div class="col-md-12"> 1299 <h2>Teaching</h2> 1300 <div class="row"> 1301 <div class="col-sm-1 col-md-1">
1302 <span class="label label-info">Winter 2021</span> 1303 </div> 1304 <div class="col-sm-11 col-md-11"> 1305 <a href="http://www-sop.inria.fr/members/Francois.Bremond/MSclass/deepLearningWinterSchool21/UCA_master/index.html" class="text-info">Deep Learning for Computer Vision</a>, Lecture and TA, <a href="http://univ-cotedazur.fr/en/idex/formations-idex/data-science/" class="text-info">MSc of Data Science and Artificial Intelligence</a>, 3IA Cote d'Azur 1306 </div> 1307 1308 <div class="col-sm-1 col-md-1"> 1309 <span class="label label-info">Winter 2020</span> 1310 </div> 1311 <div class="col-sm-11 col-md-11"> 1312 <a href="http://www-sop.inria.fr/members/Francois.Bremond/MSclass/deepLearningWinterSchool/UCA_master/index.html" class="text-info">Deep Learning for Computer Vision</a>, Lecture and TA, <a href="http://univ-cotedazur.fr/en/idex/formations-idex/data-science/" class="text-info">MSc of Data Science and Artificial Intelligence</a>, 3IA Cote d"Azur 1313 </div> 1314 </div> 1315 1316 </div> 1317 </div> <!-- end of teaching --> 1318 1319 1320 </div> <!-- end of contents --> 1321 <hr> 1322 1323 1324 <div class="container"> 1325 <footer> 1326 <p align="right"><small>Copyright © Yaohui Wang</small></p> 1327 </footer> 1328 <div style="height:10px;"></div> 1329 </div> 1330 1331 <!-- Bootstrap core JavaScript --> 1332 <!-- Placed at the end of the document so the pages load faster --> 1333
1333<script src="jq/jquery-1.11.1.min.js"></script>
1333 1334
1334<script src="js/bootstrap.min.js"></script>
1334 1335
1335<script src="js/docs.min.js"></script>
1335 1336 1337 </body> 1338</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.