Spaces:
Running
Running
File size: 13,228 Bytes
c1db41d |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 |
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>JarvisArt: Tech Poster</title>
<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.4.0/css/all.min.css">
<style>
:root {
--primary: #6e48aa;
--secondary: #9d50bb;
--accent: #4776e6;
--dark: #1a1a2e;
--light: #f8f9fa;
--text: #333;
}
* {
margin: 0;
padding: 0;
box-sizing: border-box;
font-family: 'Segoe UI', Tahoma, Geneva, Verdana, sans-serif;
}
body {
background-color: var(--dark);
color: var(--light);
line-height: 1.6;
}
.poster {
max-width: 1200px;
margin: 2rem auto;
background: linear-gradient(135deg, #1a1a2e 0%, #16213e 100%);
border-radius: 15px;
overflow: hidden;
box-shadow: 0 20px 40px rgba(0, 0, 0, 0.3);
position: relative;
border: 1px solid rgba(255, 255, 255, 0.1);
}
.poster::before {
content: '';
position: absolute;
top: 0;
left: 0;
width: 100%;
height: 100%;
background:
radial-gradient(circle at 20% 30%, rgba(110, 72, 170, 0.15) 0%, transparent 30%),
radial-gradient(circle at 80% 70%, rgba(157, 80, 187, 0.15) 0%, transparent 30%);
pointer-events: none;
}
.header {
padding: 3rem 4rem 2rem;
background: linear-gradient(to right, var(--primary), var(--secondary));
position: relative;
overflow: hidden;
}
.header::after {
content: '';
position: absolute;
bottom: -50px;
left: 0;
width: 100%;
height: 100px;
background: var(--dark);
transform: skewY(-3deg);
z-index: 1;
}
.title {
font-size: 3.5rem;
font-weight: 800;
margin-bottom: 1rem;
line-height: 1.1;
text-shadow: 2px 2px 4px rgba(0, 0, 0, 0.3);
position: relative;
z-index: 2;
}
.subtitle {
font-size: 1.5rem;
font-weight: 300;
opacity: 0.9;
margin-bottom: 1.5rem;
position: relative;
z-index: 2;
}
.authors {
display: flex;
flex-wrap: wrap;
gap: 0.5rem 1.5rem;
margin-bottom: 1rem;
position: relative;
z-index: 2;
}
.author {
font-size: 1rem;
font-weight: 500;
}
.affiliations {
display: flex;
flex-wrap: wrap;
gap: 0.5rem 1.5rem;
font-size: 0.9rem;
opacity: 0.8;
margin-bottom: 1rem;
position: relative;
z-index: 2;
}
.project-link {
display: inline-flex;
align-items: center;
gap: 0.5rem;
background: rgba(255, 255, 255, 0.1);
padding: 0.5rem 1rem;
border-radius: 50px;
text-decoration: none;
color: white;
font-weight: 500;
transition: all 0.3s ease;
position: relative;
z-index: 2;
}
.project-link:hover {
background: rgba(255, 255, 255, 0.2);
transform: translateY(-2px);
}
.content {
display: grid;
grid-template-columns: 1fr 1fr;
gap: 2rem;
padding: 3rem 4rem;
position: relative;
z-index: 2;
}
.abstract {
grid-column: 1 / -1;
background: rgba(255, 255, 255, 0.05);
padding: 2rem;
border-radius: 10px;
border-left: 4px solid var(--accent);
}
.abstract-title {
font-size: 1.5rem;
margin-bottom: 1rem;
color: var(--accent);
}
.abstract-text {
font-size: 1.1rem;
line-height: 1.7;
}
.highlight {
color: var(--accent);
font-weight: 600;
}
.features {
display: grid;
grid-template-columns: 1fr;
gap: 1.5rem;
}
.feature {
background: rgba(255, 255, 255, 0.05);
padding: 1.5rem;
border-radius: 10px;
transition: all 0.3s ease;
}
.feature:hover {
transform: translateY(-5px);
box-shadow: 0 10px 20px rgba(0, 0, 0, 0.2);
}
.feature-title {
font-size: 1.2rem;
margin-bottom: 0.5rem;
color: var(--accent);
display: flex;
align-items: center;
gap: 0.5rem;
}
.feature-icon {
font-size: 1.5rem;
}
.feature-text {
font-size: 1rem;
opacity: 0.9;
}
.image-container {
position: relative;
border-radius: 10px;
overflow: hidden;
box-shadow: 0 10px 20px rgba(0, 0, 0, 0.3);
height: 100%;
display: flex;
align-items: center;
justify-content: center;
}
.image-container img {
width: 100%;
height: auto;
object-fit: cover;
transition: transform 0.5s ease;
}
.image-container:hover img {
transform: scale(1.03);
}
.image-caption {
position: absolute;
bottom: 0;
left: 0;
right: 0;
background: rgba(0, 0, 0, 0.7);
padding: 1rem;
font-size: 0.9rem;
text-align: center;
}
.stats {
display: grid;
grid-template-columns: repeat(2, 1fr);
gap: 1.5rem;
margin-top: 1.5rem;
}
.stat {
background: rgba(255, 255, 255, 0.05);
padding: 1.5rem;
border-radius: 10px;
text-align: center;
}
.stat-value {
font-size: 2.5rem;
font-weight: 700;
color: var(--accent);
margin-bottom: 0.5rem;
}
.stat-label {
font-size: 0.9rem;
opacity: 0.8;
}
.footer {
padding: 2rem 4rem;
background: rgba(0, 0, 0, 0.3);
text-align: center;
font-size: 0.9rem;
opacity: 0.7;
position: relative;
z-index: 2;
}
@media (max-width: 992px) {
.content {
grid-template-columns: 1fr;
padding: 2rem;
}
.title {
font-size: 2.5rem;
}
.subtitle {
font-size: 1.2rem;
}
.header {
padding: 2rem;
}
}
@media (max-width: 576px) {
.title {
font-size: 2rem;
}
.subtitle {
font-size: 1rem;
}
.header {
padding: 1.5rem;
}
.content {
padding: 1.5rem;
}
.abstract {
padding: 1.5rem;
}
.stats {
grid-template-columns: 1fr;
}
}
</style>
</head>
<body>
<div class="poster">
<div class="header">
<h1 class="title">JarvisArt</h1>
<p class="subtitle">Liberating Human Artistic Creativity via an Intelligent Photo Retouching Agent</p>
<div class="authors">
<span class="author">Yunlong Lin*</span>
<span class="author">Zixu Lin*</span>
<span class="author">Kunjie Lin*</span>
<span class="author">Jinbin Bai</span>
<span class="author">Panwang Pan</span>
<span class="author">Chenxin Li</span>
<span class="author">Haoyu Chen</span>
<span class="author">Zhongdao Wang</span>
<span class="author">Xinghao Ding†</span>
<span class="author">Wenbo Li♣</span>
<span class="author">Shuicheng Yan†</span>
</div>
<div class="affiliations">
<span>Xiamen University</span>
<span>HKUST(GZ)</span>
<span>CUHK</span>
<span>Bytedance</span>
<span>NUS</span>
<span>Tsinghua University</span>
</div>
<a href="https://jarvisart.vercel.app/" class="project-link" target="_blank">
<i class="fas fa-external-link-alt"></i>
Project Page
</a>
</div>
<div class="content">
<div class="abstract">
<h2 class="abstract-title">Abstract</h2>
<p class="abstract-text">
We introduce <span class="highlight">JarvisArt</span>, a multi-modal large language model (MLLM)-driven agent that understands user intent, mimics professional artists' reasoning, and intelligently coordinates over <span class="highlight">200 retouching tools</span> within Lightroom. JarvisArt undergoes a two-stage training process and demonstrates <span class="highlight">user-friendly interaction</span>, superior generalization, and fine-grained control over both global and local adjustments. Notably, it outperforms GPT-4o with a <span class="highlight">60% improvement</span> in average pixel-level metrics on our MMArt-Bench benchmark while maintaining comparable instruction-following capabilities.
</p>
</div>
<div class="features">
<div class="feature">
<h3 class="feature-title">
<i class="fas fa-brain feature-icon"></i>
Professional Reasoning
</h3>
<p class="feature-text">
Mimics the reasoning process of professional artists through Chain-of-Thought supervised fine-tuning and GRPO-R optimization.
</p>
</div>
<div class="feature">
<h3 class="feature-title">
<i class="fas fa-tools feature-icon"></i>
Comprehensive Toolset
</h3>
<p class="feature-text">
Intelligently coordinates over 200 retouching tools within Lightroom for both global and local adjustments.
</p>
</div>
<div class="feature">
<h3 class="feature-title">
<i class="fas fa-user-astronaut feature-icon"></i>
User-Friendly Interaction
</h3>
<p class="feature-text">
Supports intuitive, free-form edits through natural inputs like text prompts, bounding boxes, or brushstrokes.
</p>
</div>
</div>
<div class="image-container">
<img src="https://cdn.vansin.top/papers/2506.17612/images/b1b85618f6f156005dab98b71efac19eb8379eb249721ebe9c59ebcd55ca3412.jpg" alt="JarvisArt Interface">
<div class="image-caption">
Figure 1: JarvisArt supports multi-granularity retouching through natural inputs and edits any-resolution images.
</div>
</div>
<div class="stats">
<div class="stat">
<div class="stat-value">200+</div>
<div class="stat-label">Retouching Tools</div>
</div>
<div class="stat">
<div class="stat-value">60%</div>
<div class="stat-label">Improvement over GPT-4o</div>
</div>
<div class="stat">
<div class="stat-value">55K</div>
<div class="stat-label">Training Samples</div>
</div>
<div class="stat">
<div class="stat-value">∞</div>
<div class="stat-label">Image Resolution Support</div>
</div>
</div>
</div>
<div class="footer">
JarvisArt: A new paradigm for intelligent photo retouching | https://jarvisart.vercel.app/
</div>
</div>
</body>
</html> |