File size: 14,630 Bytes
5edd223 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 |
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="description"
content="🐈 CatVTON: Concatenation Is All You Need for Virtual Try-On with Diffusion Models">
<meta name="keywords" content="">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>🐈 CatVTON: Concatenation Is All You Need for Virtual Try-On with Diffusion Models</title>
<script async src="https://www.googletagmanager.com/gtag/js?id=G-PYVRSFMDRL"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag() {
dataLayer.push(arguments);
}
gtag('js', new Date());
gtag('config', 'G-PYVRSFMDRL');
</script>
<link href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro"
rel="stylesheet">
<link rel="stylesheet" href="resource/css/bulma.min.css">
<link rel="stylesheet" href="resource/css/bulma-carousel.min.css">
<link rel="stylesheet" href="resource/css/bulma-slider.min.css">
<link rel="stylesheet" href="resource/css/fontawesome.all.min.css">
<link rel="stylesheet"
href="https://cdn.jsdelivr.net/gh/jpswalsh/academicons@1/css/academicons.min.css">
<link rel="stylesheet" href="resource/css/index.css">
<link rel="icon" href="resource/images/favicon.svg">
<script src="https://ajax.googleapis.com/ajax/libs/jquery/3.5.1/jquery.min.js"></script>
<script defer src="resource/js/fontawesome.all.min.js"></script>
<script src="resource/js/bulma-carousel.min.js"></script>
<script src="resource/js/bulma-slider.min.js"></script>
<script src="resource/js/index.js"></script>
</head>
<body>
<section class="hero">
<div class="hero-body">
<div class="container is-max-desktop">
<div class="columns is-centered">
<div class="column has-text-centered">
<h1 class="title is-1 publication-title">🐈 CatVTON: Concatenation Is All You Need for Virtual Try-On with Diffusion Models</h1>
<div class="is-size-5 publication-authors">
<span class="author-block">
<a href="">Zheng Chong</a><sup>1,3</sup>,</span>
<span class="author-block">
<a href="">Xiao Dong</a><sup>1</sup>,</span>
<span class="author-block">
<a href="">Haoxiang Li</a><sup>2</sup>,</span>
<span class="author-block">
<a href="">Shiyue Zhang</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="">Wenqing Zhang</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="">Xujie Zhang</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="">Hanqing Zhao</a><sup>3,4</sup>,
</span>
<span class="author-block">
<a href="">Xiaodan Liang</a><sup>*1,3</sup>,
</span>
</div>
<div class="is-size-5 publication-authors">
<span class="author-block"><sup>1</sup>Sun Yat-Sen University,</span>
<span class="author-block"><sup>2</sup>Pixocial Technology,</span>
<span class="author-block"><sup>3</sup>Peng Cheng Laboratory,</span>
<span class="author-block"><sup>4</sup>SIAT</span>
</div>
<div class="column has-text-centered">
<div class="publication-links">
<!-- PDF Link. -->
<span class="link-block">
<a href="https://arxiv.org/pdf/2407.15886"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-file-pdf"></i>
</span>
<span>Paper</span>
</a>
</span>
<!-- Arxiv Link. -->
<span class="link-block">
<a href="http://arxiv.org/abs/2407.15886"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="ai ai-arxiv"></i>
</span>
<span>arXiv</span>
</a>
</span>
<!-- Demo Link. -->
<span class="link-block">
<a href="http://120.76.142.206:8888"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-gamepad"></i>
</span>
<span>Demo</span>
</a>
</span>
<!-- Demo Link. -->
<span class="link-block">
<a href="https://huggingface.co/spaces/zhengchong/CatVTON"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-gamepad"></i>
</span>
<span>Space</span>
</a>
</span>
<!-- Models Link. -->
<span class="link-block">
<a href="https://huggingface.co/zhengchong/CatVTON"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-cube"></i>
</span>
<span>Models</span>
</a>
</span>
<!-- Code Link. -->
<span class="link-block">
<a href="https://github.com/Zheng-Chong/CatVTON"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fab fa-github"></i>
</span>
<span>Code</span>
</a>
</span>
</div>
</div>
</div>
</div>
</div>
</div>
</section>
<section class="hero teaser">
<div class="container is-max-desktop">
<div class="hero-body">
<img src="resource/img/teaser.jpg" alt="teaser">
<p>
CatVTON is a simple and efficient virtual try-on diffusion model with 1) Lightweight Network (899.06M parameters totally),
2) Parameter-Efficient Training (49.57M parameters trainable) and 3) Simplified Inference (< 8G VRAM for 1024X768
resolution).
</p>
</div>
</div>
</section>
<!-- Abstract -->
<section class="section">
<div class="container is-max-desktop">
<!-- Abstract. -->
<div class="columns is-centered has-text-centered">
<div class="column is-four-fifths">
<h2 class="title is-3">Abstract</h2>
<div class="content has-text-justified">
<p>
Virtual try-on methods based on diffusion models achieve realistic try-on effects but replicate the backbone network
as a ReferenceNet or leverage additional image encoders to process condition inputs, resulting in high training and
inference costs.
In this work, we rethink the necessity of ReferenceNet and image encoders and innovate the interaction between garment
and person, proposing CatVTON, a simple and efficient virtual try-on diffusion model. It facilitates the seamless
transfer of in-shop or worn garments of arbitrary categories to target persons by simply concatenating them in spatial
dimensions as inputs. The efficiency of our model is demonstrated in three aspects:
(1) Lightweight network. Only the original diffusion modules are used, without additional network modules. The text
encoder and cross attentions for text injection in the backbone are removed, further reducing the parameters by 167.02M.
(2) Parameter-efficient training. We identified the try-on relevant modules through experiments and achieved
high-quality try-on effects by training only 49.57M parameters (~5.51% of the backbone network’s parameters).
(3) Simplified inference. CatVTON eliminates all unnecessary conditions and preprocessing steps, including
pose estimation, human parsing, and text input, requiring only garment reference, target person image, and mask for
the virtual try-on process.
Extensive experiments demonstrate that CatVTON achieves superior qualitative and
quantitative results with fewer prerequisites and trainable parameters than baseline methods. Furthermore,
CatVTON shows good generalization in in-the-wild scenarios despite using open-source datasets with only 73K samples.
</p>
</div>
</div>
</div>
<!--/ Abstract. -->
</div>
</section>
<section class="section">
<div class="container is-max-desktop">
<!-- Architecture. -->
<div class="columns is-centered">
<div class="column is-full-width">
<h2 class="title is-3">Architecture</h2>
<div class="content has-text-justified">
<img src="resource/img/architecture.jpg">
<p>
Our method achieves the high-quality try-on by simply concatenating the conditional image (garment or reference person)
with the target person image in the spatial dimension, ensuring they remain in the same feature space throughout the
diffusion process. Only the self-attention parameters, which provide global interaction, are learnable during training.
Unnecessary cross-attention for text interaction is omitted, and no additional conditions, such as pose and parsing,
are required. These factors result in a lightweight network with minimal trainable parameters and simplified inference.
</p>
</div>
</div>
</div>
<!-- Two Columns -->
<div class="columns is-centered">
<!-- Visual Effects. -->
<div class="column">
<div class="content">
<h2 class="title is-3">Structure Comparison</h2>
<p>
We illustrate simple structure comparison of different kinds of try-on methods below. Our approach neither relies on warped garments nor
requires the heavy ReferenceNet for additional garment encoding; it only needs simple concatenation of the garment
and person images as input to obtain high-quality try-on results.
</p>
<img src="resource/img/structure.jpg">
</div>
</div>
<!-- Efficiency Comparison -->
<div class="column">
<h2 class="title is-3">Efficiency Comparison</h2>
<div class="columns is-centered">
<div class="column content">
<p>
We represent each method by two concentric circles,
where the outer circle denotes the total parameters and the inner circle denotes the trainable parameters, with the
area proportional to the parameter number. CatVTON achieves lower FID on the VITONHD dataset with fewer total
parameters, trainable parameters, and memory usage.
</p>
<img src="resource/img/efficency.jpg">
</div>
</div>
</div>
</div>
<!-- Demo -->
<div class="columns is-centered">
<div class="column is-full-width">
<h2 class="title is-3">Online Demo</h2>
<div class="content has-text-justified">
<!-- <iframe src="http://120.76.142.206:8888" width="100%" height="700px" frameborder="1/0" name="demo" scrolling="yes/no/auto">
</iframe> -->
<p>
Since GitHub Pages does not support embedded web pages, please jump to our <a href="http://120.76.142.206:8888">Demo </a>.
</p>
</div>
</div>
</div>
<!-- Acknowledgement -->
<div class="columns is-centered">
<div class="column is-full-width">
<h2 class="title is-3">Acknowledgement</h2>
<div class="content has-text-justified">
<p>
Our code is modified based on <a href="https://github.com/huggingface/diffusers">Diffusers</a>.
We adopt <a href="https://huggingface.co/runwayml/stable-diffusion-inpainting">Stable Diffusion v1.5 inpainitng</a> as base model.
We use <a href="https://github.com/GoGoDuck912/Self-Correction-Human-Parsing/tree/master">SCHP</a>
and <a href="https://github.com/facebookresearch/DensePose">DensePose</a> to automatically generate masks in our
<a href="https://github.com/gradio-app/gradio">Gradio</a> App.
Thanks to all the contributors!
</p>
</div>
</div>
</div>
<!-- "BibTeX -->
<div class="container is-max-desktop content">
<h2 class="title">BibTeX</h2>
<pre><code>
@misc{chong2024catvtonconcatenationneedvirtual,
title={CatVTON: Concatenation Is All You Need for Virtual Try-On with Diffusion Models},
author={Zheng Chong and Xiao Dong and Haoxiang Li and Shiyue Zhang and Wenqing Zhang and Xujie Zhang and Hanqing Zhao and Xiaodan Liang},
year={2024},
eprint={2407.15886},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2407.15886},
}
</code></pre>
</div>
</div>
</section>
<footer class="footer">
<div class="container">
<div class="content has-text-centered">
<a class="icon-link" href="http://arxiv.org/abs/2407.15886" class="external-link" disabled>
<i class="ai ai-arxiv"></i>
</a>
<a class="icon-link" href="https://arxiv.org/pdf/2407.15886">
<i class="fas fa-file-pdf"></i>
</a>
<a class="icon-link" href="http://120.76.142.206:8888" class="external-link" disabled>
<i class="fas fa-gamepad"></i>
</a>
<a class="icon-link" href="https://github.com/Zheng-Chong/CatVTON" class="external-link" disabled>
<i class="fab fa-github"></i>
</a>
<a class="icon-link" href="https://huggingface.co/zhengchong/CatVTON" class="external-link" disabled>
<i class="fas fa-cube"></i>
</a>
</div>
<div class="columns is-centered">
<div class="column is-8">
<div class="content">
<p>
This website is modified from <a href="https://nerfies.github.io/">Nerfies</a>. Thanks for the great work!
Their source code is available on <a href="https://github.com/nerfies/nerfies.github.io">GitHub</a>.
</p>
</div>
</div>
</div>
</div>
</footer>
</body>
</html> |