AISmartPerception.github.io/index.html at main · AISmartPerception/AISmartPerception.github.io · GitHub

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
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
<!doctype html>
<html lang="en">
<head>
  <meta charset="utf-8" />
  <meta name="viewport" content="width=device-width,initial-scale=1" />
  <title>AI Smart Perception — AISmartPerception</title>

<meta name="description" content="AI Smart Perception – Research in Computer Vision, Multimodal Large Models, Perceptual Quality, ISP Tuning, and Optimization.">
<meta name="keywords" content="Computer Vision, Multimodal Models, Image Quality Assessment, ISP, Reinforcement Learning, Optimization">
<meta name="author" content="AISmartPerception">
<meta name="description" content="AI Smart Perception team homepage: projects, research areas, and links." />

  <style>
    :root{
      --bg0:#070A12;
      --bg1:#0B1020;
      --card: rgba(255,255,255,.06);
      --card2: rgba(255,255,255,.08);
      --stroke: rgba(255,255,255,.12);
      --text: rgba(255,255,255,.92);
      --muted: rgba(255,255,255,.68);
      --shadow: 0 20px 60px rgba(0,0,0,.45);
      --radius: 20px;
      --radius2: 16px;
      --accentA:#7C3AED; /* purple */
      --accentB:#06B6D4; /* cyan */
      --accentC:#F97316; /* orange */
      --accentD:#22C55E; /* green */
      --accentE:#F43F5E; /* rose */
      --link: #A5B4FC;
    }

    *{ box-sizing:border-box; }
    html,body{ height:100%; }
    body{
      margin:0;
      font-family: ui-sans-serif, system-ui, -apple-system, Segoe UI, Roboto, Helvetica, Arial, "Apple Color Emoji","Segoe UI Emoji";
      color: var(--text);
      background: radial-gradient(1200px 800px at 20% -10%, rgba(124,58,237,.35), transparent 60%),
                  radial-gradient(1000px 700px at 110% 20%, rgba(6,182,212,.25), transparent 60%),
                  radial-gradient(900px 600px at 40% 110%, rgba(244,63,94,.18), transparent 60%),
                  linear-gradient(180deg, var(--bg0), var(--bg1));
      overflow-x:hidden;
    }

    a{ color: var(--link); text-decoration:none; }
    a:hover{ text-decoration:underline; }

    .container{
      width:min(1100px, calc(100% - 32px));
      margin: 0 auto;
      padding: 28px 0 60px;
    }

    /* Top nav */
    .nav{
      display:flex;
      align-items:center;
      justify-content:space-between;
      gap:16px;
      padding: 10px 14px;
      border: 1px solid var(--stroke);
      background: rgba(0,0,0,.18);
      border-radius: 999px;
      backdrop-filter: blur(10px);
      box-shadow: 0 12px 40px rgba(0,0,0,.25);
    }
    .brand{
      display:flex; align-items:center; gap:10px;
      padding-left:6px;
      font-weight:700;
      letter-spacing:.2px;
    }
    .logo{
      width:34px; height:34px; border-radius: 12px;
      background:
        radial-gradient(circle at 30% 30%, rgba(255,255,255,.35), transparent 45%),
        linear-gradient(135deg, var(--accentA), var(--accentB));
      box-shadow: 0 10px 30px rgba(124,58,237,.25);
      border: 1px solid rgba(255,255,255,.2);
    }
    .navlinks{
      display:flex; gap:14px; flex-wrap:wrap;
      padding-right:6px;
      color: var(--muted);
      font-size: 14px;
    }
    .navlinks a{ color: var(--muted); }
    .navlinks a:hover{ color: var(--text); text-decoration:none; }

    /* Hero */
    .hero{
      margin-top: 22px;
      padding: 34px 26px;
      border: 1px solid var(--stroke);
      background:
        radial-gradient(900px 500px at 10% 0%, rgba(124,58,237,.25), transparent 55%),
        radial-gradient(900px 500px at 90% 30%, rgba(6,182,212,.18), transparent 55%),
        rgba(255,255,255,.04);
      border-radius: var(--radius);
      box-shadow: var(--shadow);
      position: relative;
      overflow:hidden;
    }
    .hero:before{
      content:"";
      position:absolute; inset:-2px;
      background: linear-gradient(120deg, rgba(124,58,237,.18), rgba(6,182,212,.12), rgba(244,63,94,.10));
      filter: blur(16px);
      opacity:.6;
      pointer-events:none;
    }
    .hero-inner{ position:relative; display:grid; gap:18px; }
    .kicker{
      display:inline-flex; align-items:center; gap:10px;
      font-size: 13px;
      color: var(--muted);
      letter-spacing:.3px;
      text-transform: uppercase;
    }
    .dot{
      width:10px; height:10px; border-radius:999px;
      background: linear-gradient(135deg, var(--accentC), var(--accentE));
      box-shadow: 0 0 0 4px rgba(249,115,22,.15);
    }
    h1{
      margin:0;
      font-size: clamp(28px, 4vw, 44px);
      line-height: 1.12;
      letter-spacing: -0.6px;
    }
    .subtitle{
      margin:0;
      color: var(--muted);
      font-size: 16px;
      line-height: 1.6;
      max-width: 68ch;
    }
    .cta-row{
      display:flex; gap:12px; flex-wrap:wrap;
      margin-top: 6px;
    }
    .btn{
      display:inline-flex; align-items:center; justify-content:center; gap:10px;
      padding: 11px 14px;
      border-radius: 999px;
      border: 1px solid var(--stroke);
      background: rgba(255,255,255,.06);
      color: var(--text);
      font-weight: 600;
      font-size: 14px;
      text-decoration:none !important;
      transition: transform .12s ease, background .12s ease, border-color .12s ease;
      user-select:none;
    }
    .btn:hover{ transform: translateY(-1px); background: rgba(255,255,255,.09); border-color: rgba(255,255,255,.18); }
    .btn-primary{
      border-color: rgba(124,58,237,.40);
      background: linear-gradient(135deg, rgba(124,58,237,.75), rgba(6,182,212,.45));
      box-shadow: 0 18px 40px rgba(124,58,237,.18);
    }
    .btn-primary:hover{ background: linear-gradient(135deg, rgba(124,58,237,.85), rgba(6,182,212,.52)); }

    /* Section titles */
    .section{
      margin-top: 22px;
      display:grid;
      gap: 14px;
    }
    .section-title{
      display:flex; align-items:flex-end; justify-content:space-between; gap: 10px;
      margin-top: 10px;
    }
    .section-title h2{
      margin:0;
      font-size: 18px;
      letter-spacing: -.2px;
    }
    .section-title p{
      margin:0;
      color: var(--muted);
      font-size: 13px;
    }

    /* Chips */
    .chips{
      display:flex;
      flex-wrap:wrap;
      gap: 10px;
    }
    .chip{
      display:inline-flex;
      align-items:center;
      gap:10px;
      padding: 10px 12px;
      border-radius: 999px;
      border: 1px solid var(--stroke);
      background: rgba(255,255,255,.05);
      color: rgba(255,255,255,.84);
      font-size: 14px;
      line-height: 1;
      box-shadow: 0 8px 24px rgba(0,0,0,.18);
    }
    .chip .ic{
      width:22px;height:22px;border-radius:8px;
      display:grid;place-items:center;
      border:1px solid rgba(255,255,255,.18);
      background: rgba(0,0,0,.18);
      font-size: 14px;
    }

    /* Cards grid */
    .grid{
      display:grid;
      gap: 14px;
      grid-template-columns: repeat(12, 1fr);
    }
    .card{
      grid-column: span 6;
      border: 1px solid var(--stroke);
      border-radius: var(--radius2);
      background: linear-gradient(180deg, rgba(255,255,255,.06), rgba(255,255,255,.03));
      box-shadow: 0 16px 44px rgba(0,0,0,.28);
      padding: 16px;
      position:relative;
      overflow:hidden;
    }
    .card:before{
      content:"";
      position:absolute;
      inset:-2px;
      background: radial-gradient(350px 220px at 30% 0%, rgba(6,182,212,.18), transparent 60%),
                  radial-gradient(350px 220px at 90% 30%, rgba(124,58,237,.18), transparent 60%),
                  radial-gradient(350px 220px at 60% 110%, rgba(249,115,22,.10), transparent 60%);
      opacity:.85;
      pointer-events:none;
    }
    .card-inner{ position:relative; display:flex; flex-direction:column; gap: 10px; }

    .card-top{ display:flex; align-items:flex-start; justify-content:space-between; gap: 12px; }
    .card-title{
      display:flex; align-items:center; gap: 10px;
      font-weight: 750;
      letter-spacing: -.2px;
      font-size: 16px;
      margin:0;
    }
    .badge{
      display:inline-flex; align-items:center; gap:8px;
      padding: 8px 10px;
      border-radius: 999px;
      border: 1px solid rgba(255,255,255,.14);
      background: rgba(0,0,0,.18);
      color: var(--muted);
      font-size: 12px;
      white-space:nowrap;
    }
    .icon{
      width:34px; height:34px; border-radius: 12px;
      border: 1px solid rgba(255,255,255,.18);
      display:grid; place-items:center;
      box-shadow: 0 12px 26px rgba(0,0,0,.25);
      background: rgba(0,0,0,.18);
      font-size: 18px;
    }
    .desc{
      margin:0;
      color: var(--muted);
      font-size: 14px;
      line-height: 1.55;
    }
    .card-actions{
      display:flex;
      gap:10px;
      flex-wrap:wrap;
      margin-top: 2px;
    }
    .btn-sm{
      padding: 9px 12px;
      font-size: 13px;
    }

    /* Responsive */
    @media (max-width: 880px){
      .card{ grid-column: span 12; }
      .nav{ border-radius: 18px; }
    }

    /* Footer */
    .footer{
      margin-top: 26px;
      padding-top: 18px;
      border-top: 1px solid rgba(255,255,255,.10);
      color: var(--muted);
      font-size: 13px;
      display:flex;
      justify-content:space-between;
      gap: 10px;
      flex-wrap:wrap;
    }
  </style>
</head>

<body>
  <div class="container">
    <header class="nav">
      <div class="brand">
        <div class="logo" aria-hidden="true"></div>
        <div>
          AI Smart Perception
          <div style="font-size:12px;color:var(--muted);font-weight:600;margin-top:2px;">AISmartPerception · GitHub Pages</div>
        </div>
      </div>

      <nav class="navlinks" aria-label="Top navigation">
        <a href="#projects">Projects</a>
        <a href="#areas">Research Areas</a>
        <a href="https://github.com/AISmartPerception" target="_blank" rel="noreferrer">GitHub</a>
      </nav>
    </header>

    <section class="hero">
      <div class="hero-inner">
        <div class="kicker"><span class="dot" aria-hidden="true"></span> Team homepage & research portfolio</div>
        <h1>AI for Smart Perception</h1>
        <p class="subtitle">
          We build practical perception systems at the intersection of computer vision, multimodal foundation models,
          camera ISP tuning, and data/compute-efficient optimization — with a focus on perceptual quality and robust deployment.
        </p>

        <div class="cta-row">
          <a class="btn btn-primary" href="#projects">🚀 Explore Projects</a>
          <a class="btn" href="https://github.com/AISmartPerception" target="_blank" rel="noreferrer">🐙 View GitHub Org</a>
        </div>
      </div>
    </section>

    <section id="areas" class="section">
      <div class="section-title">
        <h2>Research Areas</h2>
        <p>Core technical pillars</p>
      </div>

      <div class="chips">
        <span class="chip"><span class="ic">🖼️</span> Computer Vision</span>
        <span class="chip"><span class="ic">🧠</span> Multi-Modal Large Models</span>
        <span class="chip"><span class="ic">⚡</span> Test-Time Scaling</span>
        <span class="chip"><span class="ic">✨</span> Perceptual Quality</span>
        <span class="chip"><span class="ic">🎛️</span> Camera ISP Tuning</span>
        <span class="chip"><span class="ic">📈</span> Optimization</span>
        <span class="chip"><span class="ic">🤖</span> Reinforcement Learning</span>
      </div>
    </section>

    <section id="projects" class="section">
      <div class="section-title">
        <h2>Projects</h2>
        <p>Each project has a dedicated page under this site</p>
      </div>

      <div class="grid">
        <article class="card">
          <div class="card-inner">
            <div class="card-top">
              <div class="card-title">
                <span class="icon" aria-hidden="true">🧩</span>
                Distortion Graphs
              </div>
              <span class="badge">IQA · region-level reasoning</span>
            </div>
            <p class="desc">ICLR 2026: We introduce a region-grounded, structured representation for comparative image assessment—modeling paired images as a Distortion Graph that captures distortion type, severity, comparisons, and quality evidence at the region level.</p>
            <div class="card-actions">
              <a class="btn btn-sm btn-primary" href="./distortion-graph/">Open page →</a>
              <a class="btn btn-sm" href="https://github.com/AISmartPerception" target="_blank" rel="noreferrer">Repo (link later)</a>
            </div>
          </div>
        </article>

        <article class="card">
          <div class="card-inner">
            <div class="card-top">
              <div class="card-title">
                <span class="icon" aria-hidden="true">🛰️</span>
                Perception Program
              </div>
              <span class="badge">context · sensing · agents</span>
            </div>
            <p class="desc">CVPR 2026: A training-free, model-agnostic interface that converts dense vision-tool outputs (depth, flow, correspondence, detection, etc.) into compact, language-native “Perception Programs” that MLLMs can reliably parse and reason over—without extra tool calls or fine-tuning.</p>
            <div class="card-actions">
              <a class="btn btn-sm btn-primary" href="./perception-program/">Open page →</a>
              <a class="btn btn-sm" href="https://github.com/AISmartPerception" target="_blank" rel="noreferrer">Repo (link later)</a>
            </div>
          </div>
        </article>

        <article class="card">
          <div class="card-inner">
            <div class="card-top">
              <div class="card-title">
                <span class="icon" aria-hidden="true">🧬</span>
                ReaGEN
              </div>
              <span class="badge">generation · reasoning</span>
            </div>
            <p class="desc">CVPR 2026: Efficient, question-adaptive structured Chain-of-Thought generation for VLMs. ReaGEN uses teacher-guided evolutionary search to learn sample-specific reasoning structures, then trains a lightweight generator to produce and refine them at test time—achieving strong gains while cutting token cost dramatically.</p>
            <div class="card-actions">
              <a class="btn btn-sm btn-primary" href="./reagen/">Open page →</a>
              <a class="btn btn-sm" href="https://github.com/AISmartPerception" target="_blank" rel="noreferrer">Repo (link later)</a>
            </div>
          </div>
        </article>

        <article class="card">
          <div class="card-inner">
            <div class="card-top">
              <div class="card-title">
                <span class="icon" aria-hidden="true">🎯</span>
                ScoreNET
              </div>
              <span class="badge">LLM/VLM IQA · prompting</span>
            </div>
            <p class="desc">WACV 2026: A metric-aggregated prompting framework for boosting CLIP- and MLLM-based no-reference image quality assessment. ScoreNET ensembles diverse IQA metrics and injects their knowledge through learned soft prompts, enabling foundation models to produce more reliable, distortion-aware perceptual scores.</p>
            <div class="card-actions">
              <a class="btn btn-sm btn-primary" href="./scorenet/">Open page →</a>
              <a class="btn btn-sm" href="https://github.com/AISmartPerception" target="_blank" rel="noreferrer">Repo (link later)</a>
            </div>
          </div>
        </article>

        <article class="card">
          <div class="card-inner">
            <div class="card-top">
              <div class="card-title">
                <span class="icon" aria-hidden="true">🌈</span>
                Chameleon Tuner
              </div>
              <span class="badge">ISP 3D LUT · misaligned supervision</span>
            </div>
            <p class="desc">WACV 2026: A region-aware, multi-objective optimization framework for automatic 3D LUT tuning in subjective ISP scenarios. ChameleonTuner establishes region-level color correspondences under FoV/PoV variations and directly optimizes LUT parameters via evolutionary search, delivering controllable, interpretable, and perceptually robust ISP color calibration.</p>
            <div class="card-actions">
              <a class="btn btn-sm btn-primary" href="./chameleon-tuner/">Open page →</a>
              <a class="btn btn-sm" href="https://github.com/AISmartPerception" target="_blank" rel="noreferrer">Repo (link later)</a>
            </div>
          </div>
        </article>
      </div>
    </section>

    <footer class="footer">
      <div>© <span id="y"></span> AISmartPerception</div>
      <div>
        <a href="https://github.com/AISmartPerception" target="_blank" rel="noreferrer">GitHub</a>
        · <a href="#projects">Projects</a>
        · <a href="#areas">Research Areas</a>
      </div>
    </footer>
  </div>

  <script>
    document.getElementById("y").textContent = new Date().getFullYear();
  </script>
</body>
</html>
HTML