changer/index.html at main · hahminlew/changer · 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
<!DOCTYPE html>
<html>
<head>
  <meta charset="utf-8">
  <meta name="description"
        content="Towards High-fidelity Head Blending with Chroma Keying for Industrial Applications.">
  <meta name="keywords" content="Head Blending, Chroma Keying, Green Screen">
  <meta name="viewport" content="width=device-width, initial-scale=1">
  <title>Towards High-fidelity Head Blending with Chroma Keying for Industrial Applications</title>

  <link href="https://fonts.googleapis.com/css?family=Google+Sans|Noto+Sans|Castoro"
        rel="stylesheet">

  <link rel="stylesheet" href="./static/css/bulma.min.css">
  <link rel="stylesheet" href="./static/css/bulma-carousel.min.css">
  <link rel="stylesheet" href="./static/css/bulma-slider.min.css">
  <link rel="stylesheet" href="./static/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="./static/css/index.css">
  <link rel="icon" href="./static/images/github.svg">

  <script src="https://ajax.googleapis.com/ajax/libs/jquery/3.5.1/jquery.min.js"></script>
  <script defer src="./static/js/fontawesome.all.min.js"></script>
  <script src="./static/js/bulma-carousel.min.js"></script>
  <script src="./static/js/bulma-slider.min.js"></script>
  <script src="./static/js/index.js"></script>
  <script src="./static/js/video_comparison.js"></script>
</head>
<body>

<nav class="navbar" role="navigation" aria-label="main navigation">
  <div class="navbar-brand">
    <a role="button" class="navbar-burger" aria-label="menu" aria-expanded="false">
      <span aria-hidden="true"></span>
      <span aria-hidden="true"></span>
      <span aria-hidden="true"></span>
    </a>
  </div>
  <div class="navbar-menu">
    <div class="navbar-start" style="flex-grow: 1; justify-content: center;">
      <a class="navbar-item" href="https://hahminlew.github.io">
      <span class="icon">
          <i class="fas fa-home"></i>
      </span>
      </a>

      <div class="navbar-item has-dropdown is-hoverable">
        <a class="navbar-link">
          More Research
        </a>
        <div class="navbar-dropdown">
          <a class="navbar-item" href="https://hahminlew.github.io/geoavatar/">
            GeoAvatar
          </a>
          <a class="navbar-item" href="https://hahminlew.github.io/changer/">
            CHANGER
          </a>
        </div>
      </div>
    </div>

  </div>
</nav>

<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">Towards High-fidelity Head Blending with Chroma Keying for Industrial Applications</h1>
          <div class="is-size-5 publication-authors">
            <span class="author-block">
              <a href="https://hahminlew.github.io/">Hah Min Lew</a><sup>1*</sup></span>
            <span class="author-block">
              <a href="https://scholar.google.com/citations?user=YDQ6lwcAAAAJ">Sahng-Min Yoo</a><sup>2*</sup></span>
            <span class="author-block">
              <a href="https://sites.google.com/view/hyunwookang">Hyunwoo Kang</a><sup>3*</sup></span>
            <span class="author-block">
              <a href="https://vgi.korea.ac.kr/">Gyeong-Moon Park</a><sup>4</sup></span>
          </div>

          <div class="is-size-5 publication-authors">
            <span class="author-block"><sup>1</sup>Klleon AI Research,</span>
            <span class="author-block"><sup>2</sup>Samsung Research,</span>
            <span class="author-block"><sup>3</sup>Hyperconnect,</span>
            <span class="author-block"><sup>4</sup>Kyung Hee University</span>
          </div>
          <div class="is-size-4 publication-authors">
            <span class="author-block"><sup></sup><b>WACV 2025</b></span>
          </div>
          <div class="is-size-5 publication-authors">
            <span class="author-block">(*equal contribution, work done at Klleon AI Research)</span>
          </div>

          <div class="column has-text-centered">
            <div class="publication-links">
              <!-- PDF Link. -->
              <span class="link-block">
                <a href="https://openaccess.thecvf.com/content/WACV2025/papers/Lew_Towards_High-Fidelity_Head_Blending_with_Chroma_Keying_for_Industrial_Applications_WACV_2025_paper.pdf"
                   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 abstract Link -->
              <span class="link-block">
                <a href="https://arxiv.org/abs/2411.00652" target="_blank"
                  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>

            </div>

          </div>
        </div>
      </div>
    </div>
  </div>
</section>

<section class="hero teaser">
  <div class="container is-max-desktop">
    <div class="hero-body">
      <video id="teaser" autoplay muted loop playsinline height="100%">
        <source src="./static/videos/sam_with_audio.mp4"
                type="video/mp4">
      </video>
      <!-- Mute/Unmute Icon Button -->
      <div id="muteToggle" style="
          position: absolute;
          bottom: 145px;
          right: 40px;
          background-color: rgba(0, 0, 0, 0.7);
          border-radius: 10%;
          padding: 10px;
          cursor: pointer;
          z-index: 10; /* Ensure it appears above the canvas */
      ">
          <img id="muteIcon" src="./static/images/mute.png" alt="Mute" style="width: 30px; height: 30px;">
      </div>
      <script>
        document.addEventListener('DOMContentLoaded', () => {
          const video = document.getElementById('teaser');
          const muteToggle = document.getElementById('muteToggle');
          const muteIcon = document.getElementById('muteIcon');

          muteToggle.addEventListener('click', () => {
            if (video.muted) {
              video.muted = false;
              muteIcon.src = './static/images/high-volume.png'; // Unmute icon
            } else {
              video.muted = true;
              muteIcon.src = './static/images/mute.png'; // Mute icon
            }
          });

          // Sync icon on volume change
          video.addEventListener('volumechange', () => {
            muteIcon.src = video.muted
              ? './static/images/mute.png'
              : './static/images/high-volume.png';
          });
        });
      </script>
      <h2 class="content has-text-justified">
        We propose <b>CHANGER</b>, a novel pipeline for <b>C</b>onsistent <b>H</b>ead blending with predictive <b>A</b>tte<b>N</b>tion <b>G</b>uided foreground <b>E</b>stimation under chroma key setting for indust<b>R</b>ial applications.
        CHANGER generates a high-fidelity head blended result.
      </h2>
    </div>
  </div>
</section>


<section class="hero is-light is-small">
  <div class="hero-body">
    <div class="container">
      <div id="results-carousel" class="carousel results-carousel">
        <div class="item item-lincoln-2">
          <video poster="" id="lincoln-2-green" autoplay controls muted loop playsinline height="100%">
            <source src="./static/videos/lincoln2_green.mp4"
                    type="video/mp4">
          </video>
        </div>
        <div class="item item-sejong-2">
          <video poster="" id="sejong-2-green" autoplay controls muted loop playsinline height="100%">
            <source src="./static/videos/sejong2_green.mp4"
                    type="video/mp4">
          </video>
        </div>
        <div class="item item-lincoln-1">
          <video poster="" id="lincoln-1-green" autoplay controls muted loop playsinline height="100%">
            <source src="./static/videos/lincoln1_green.mp4"
                    type="video/mp4">
          </video>
        </div>
        <div class="item item-sejong-1">
          <video poster="" id="sejong-1-green" autoplay controls muted loop playsinline height="100%">
            <source src="./static/videos/sejong1_green.mp4"
                    type="video/mp4">
          </video>
        </div>
        <div class="item item-lincoln-3">
          <video poster="" id="lincoln-3-green" autoplay controls muted loop playsinline height="100%">
            <source src="./static/videos/lincoln3_green.mp4"
                    type="video/mp4">
          </video>
        </div>
        <div class="item item-sejong-3">
          <video poster="" id="sejong-3-green" autoplay controls muted loop playsinline height="100%">
            <source src="./static/videos/sejong3_green.mp4"
                    type="video/mp4">
          </video>
        </div>
      </div>
    </div>
    <h2 class="subtitle has-text-centered">
      Head blending results from two source images to three target videos.
    </h2>
  </div>
</section>

<hr>

<section class="hero teaser">
  <div class="container is-max-desktop">
    <div class="hero-body">
      <div class="columns is-centered has-text-centered">
        <div class="column is-four-fifths">
          <h2 class="title is-3">High-Fidelity Industrial Head Blending Pipeline</h2>
          <img src="./static/images/pipeline.jpg" class="center">
          <h2 class="content has-text-justified">
            Illustration of our CHANGER pipeline: After acquiring the actor's frames (source), we can seamlessly insert acting scenes into the desired scenes with our CHANGER.
            Chroma keying ensures high-fidelity backgrounds. Here, both of the source and the target actors are virtual humans.
          </h2>
        </div>
      </div>
    </div>
  </div>
</section>

<section class="hero teaser">
  <div class="container is-max-desktop">
    <div class="hero-body">
      <div class="columns is-centered has-text-centered">
        <div class="column is-four-fifths">
          <h2 class="title is-3">Various Industrial Application Examples</h2>
          <img src="./static/images/industrial_examples.jpg" class="center">
          <h2 class="content has-text-justified">
            By leveraging chroma key technique with our proposed CHANGER pipeline, we can obtain various high-fidelity head blended videos in the wild environments.
            The red boxes represent the source images.
          </h2>
        </div>
      </div>
    </div>
  </div>
</section>

<hr>

<section class="section">
  <div class="container is-max-desktop">
    <div class="columns is-centered has-text-centered">
      <div class="column is-four-fifths">
        <h2 class="title is-3">Effects of Our Contributions</h2>
        <div class="content has-text-justified">
          <p>
            Our novel H<sup>2</sup> augmentation and FPAT with chroma keying successfully achieves high-fidelity head blending performance.
            Point and drag your mouse to see the results!
          </p>
        </div>
      </div>
    </div>
    <div class="container is-max-desktop">
      <div class="columns is-centered">
          <div class="video-compare-container">
              <video class="video" id="lincoln1" loop playsinline autoPlay muted src="./static/videos/lincoln1.mp4" onplay="resizeAndPlay(this)"></video>
              <canvas height=0 class="videoMerge" id="lincoln1Merge"></canvas>
          </div>
      </div>
    </div>
    <div class="container is-max-desktop">
      <div class="columns is-centered">
          <div class="video-compare-container">
              <video class="video" id="sejong1" loop playsinline autoPlay muted src="./static/videos/sejong1.mp4" onplay="resizeAndPlay(this)"></video>
              <canvas height=0 class="videoMerge" id="sejong1Merge"></canvas>
          </div>
      </div>
    </div>
  </div>
</section>

<hr>

<section class="section">
  <div class="container is-max-desktop">
    <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>
            We introduce industrial Head Blending pipeline for the task of seamlessly integrating an actor’s head onto a target body in digital content creation, which is essential for applications such as visual effects (VFX), digital human creation, and virtual avatars.
            The key challenge stems from discrepancies in head shape and hair structure, which lead to unnatural boundaries and blending artifacts.
            Existing methods treat foreground and background as a single task, resulting in suboptimal blending quality.
            To address this problem, we propose <b>CHANGER</b>, a novel pipeline that decouples background integration from foreground blending.
            By utilizing chroma keying for artifact-free background generation and introducing <b>H</b>ead shape and long <b>H</b>air <b>augmentation</b> (<b>H<sup>2</sup> augmentation</b>) to simulate a wide range of head shapes and hair styles, CHANGER improves generalization on innumerable various real-world cases.
            Furthermore, our <b>F</b>oreground <b>P</b>redictive <b>A</b>ttention <b>T</b>ransformer (<b>FPAT</b>) module enhances foreground blending by predicting and focusing on key head and body regions.
            Quantitative and qualitative evaluations on benchmark datasets demonstrate that our CHANGER outperforms state-of-the-art methods, delivering high-fidelity, industrial-grade results.
          </p>
        </div>
      </div>
    </div>
  </div>
</section>

<section class="hero teaser">
  <div class="container is-max-desktop">
    <div class="hero-body">
      <div class="columns is-centered has-text-centered">
        <div class="column is-four-fifths">
          <h2 class="title is-3">Motivations</h2>
          <img src="./static/images/motivation.jpg" class="center">
          <h2 class="content has-text-justified">
            We propose CHANGER to consider the real-world application. As shown in (a), the existing work [2] (<a href="">H2SB</a>) shows severe artifacts on inpainting regions.
            To inpaint the background flawlessly, we propose to introduce chroma keying in the head blending framework.
            However, it still shows low-fidelity results to inpaint the neck and the body, which is hidden due to the head shape and hair difference described in a red box of (b).
            CHANGER generates the high-fidelity foreground with H<sup>2</sup> augmentation and Foreground Predictive Attention Transformer (FPAT).
            CHANGER removes artifacts as shown in the blue boxes of (b) and (c), and easily changes various high-fidelity real-world backgrounds.
          </h2>
        </div>
      </div>
    </div>
  </div>
</section>

<section class="hero teaser">
  <div class="container is-max-desktop">
    <div class="hero-body">
      <div class="columns is-centered has-text-centered">
        <div class="column is-four-fifths">
          <h2 class="title is-3">Methods</h2>
          <img src="./static/images/network.jpg" class="center">
          <h2 class="content has-text-justified">
            CHANGER consists of H<sup>2</sup> augmentation, an encoder (E), a head colorizer, a body blender including Foreground Predictive Attention Transformer (FPAT) modules, and a decoder (D).
            We visualize our proposed input (X) manipulation method with overall training (blue) and inference (red) schemes.
            The head colorizer colorizes the gray head of X, and the body blender inpaints the hidden body and the neck with a foreground mask-aware attention mechanism.
            The Foreground-Prediction module predicts the foreground mask (M) of the body and the neck region, and the attention is reweighted according to M.
          </h2>
        </div>
      </div>
    </div>
  </div>
</section>

<section class="hero teaser">
  <div class="container is-max-desktop">
    <div class="hero-body">
      <div class="columns is-centered has-text-centered">
        <div class="column is-four-fifths">
          <h2 class="title is-3">Comparisons</h2>
          <img src="./static/images/main_comparison.jpg" class="center">
          <h2 class="content has-text-justified">
            We compared qualitative results between CHANGER with the state-of-the-art head blending and face swapping models: H2SB [2], SimSwap [3], RAFSwap [4], BlendFace [5], and E4S [6].
            "CK" represents inference on the chromakey configuration.
          </h2>
        </div>
      </div>
    </div>
  </div>
</section>

<section class="section">
  <div class="container is-max-desktop">

    <div class="columns is-centered">

    <!-- Concurrent Work. -->
    <div class="columns is-centered">
      <div class="column is-full-width">
        <h2 class="title is-3">References</h2>

        <div class="content has-text-justified">
          <p>
            We utilized the pre-trained face reenactment network to generate our head blending results.
          </p>
          <p>
            [1] <a href="https://github.com/wyhsirius/LIA">Latent Image Animator: Learning to Animate Images via Latent Space Navigation</a>.
          </p>
          <p>
            [2] <a href="https://github.com/jmliu88/HeSer">Few-Shot Head Swapping in the Wild</a>.
          </p>
          <p>
            [3] <a href="https://github.com/neuralchen/SimSwap">SimSwap: An Efficient Framework For High Fidelity Face Swapping</a>.
          </p>
          <p>
            [4] <a href="https://github.com/xc-csc101/RAFSwap">Region-Aware Face Swapping</a>.
          </p>
          <p>
            [5] <a href="https://github.com/mapooon/BlendFace">BlendFace: Re-designing Identity Encoders for Face-Swapping</a>.
          </p>
          <p>
            [6] <a href="https://github.com/e4s2022/e4s">E4S: Fine-grained Face Swapping via Regional GAN Inversion</a>.
          </p>
        </div>
      </div>
    </div>
    <!--/ Concurrent Work. -->

  </div>
</section>


<section class="section" id="BibTeX">
  <div class="container is-max-desktop content">
    <h2 class="title">BibTeX</h2>
    <pre><code>@inproceedings{lew2025towards,
      title={Towards High-fidelity Head Blending with Chroma Keying for Industrial Applications},
      author={Lew, Hah Min and Yoo, Sahng-Min and Kang, Hyunwoo and Park, Gyeong-Moon},
      booktitle={2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
      pages={6188--6196},
      year={2025},
      organization={IEEE}
    }</code></pre>
  </div>
</section>


<footer class="footer">
  <div class="container">
    <div class="content has-text-centered">
      <a class="icon-link"
         href="https://arxiv.org/abs/2411.00652">
        <i class="fas fa-file-pdf"></i>
      </a>
    </div>
    <div class="columns is-centered">
      <div class="column is-8">
        <div class="content">
          <p>
            This website including source images and backgrounds is licensed under a <a rel="license"
                                                href="http://creativecommons.org/licenses/by-sa/4.0/">Creative
            Commons Attribution-ShareAlike 4.0 International License</a>.
          </p>
          <p>
            This project page is borrowed from the <a
              href="https://github.com/nerfies/nerfies.github.io">Nerfies Authors</a>,
            and partially employ the point-and-drag function via mouse from <a
            href="https://dorverbin.github.io/refnerf/">Ref-Nerf</a>.
            Thanks for sharing the awesome templates.
          </p>
      </div>
    </div>
  </div>
</footer>

</body>
</html>