-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
404 lines (374 loc) · 15.8 KB
/
index.html
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
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="description"
content="This is the supplementary webpage for CVPR submission -- Diffusion Priors for Dynamic Novel View Synthesis from Monocular Videos.">
<meta name="keywords" content="Nerfies, D-NeRF, NeRF, Dynamic novel view synthesis, dpdy, DpDy">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Diffusion Priors for Dynamic View Synthesis Using Monocular Videos</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/favicon.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>
</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">Diffusion Priors for Dynamic View Synthesis from Monocular Videos </h1>
<div class="is-size-5 publication-authors">
<span class="author-block">
<a href="https://mightychaos.github.io/">Chaoyang Wang</a><sup>1</sup>,</span>
<span class="author-block">
<a href="https://payeah.net/">Peiye Zhuang</a><sup>1</sup>,</span>
<span class="author-block">
<a href="https://aliaksandrsiarohin.github.io/aliaksandr-siarohin-website/">Aliaksandr Siarohin</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="https://research.snap.com/team/team-member.html#junli-cao">Junli Cao</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="https://guochengqian.github.io/">Guocheng Qian</a><sup>1,2</sup>,
</span>
<span class="author-block">
<a href="http://hsinyinglee.com/">Hsin-Ying Lee</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="http://www.stulyakov.com/">Sergey Tulyakov</a><sup>1</sup>,
</span>
</div>
<div class="is-size-5 publication-authors">
<span class="author-block"><sup>1</sup>Snap Inc.,</span>
<span class="author-block"><sup>2</sup>KAUST</span>
</div>
</div>
</div>
</div>
</div>
</section>
<table align=center width=1000px>
<!-- <center><h2 class="title is-3">Novel view + depth map</h2></center>
<tr>
<center>
<div style='width:90%; text-align:justify; float:center'>
</center>
</tr> -->
<tr>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:300px' src='./resources/demos/ours-backpack.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:300px' src='./resources/demos/ours-creeper.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:300px' src='./resources/demos/ours-spin.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:300px' src='./resources/demos/ours-mochi-high-five.mp4'>
</center>
</td>
</tr>
<tr>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:300px' src='./resources/demos/ours-pillow.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:300px' src='./resources/demos/ours-block.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:300px' src='./resources/demos/ours-sriracha-tree.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:300px' src='./resources/demos/ours-wheel.mp4'>
</center>
</td>
</tr>
</table>
<section class="section">
<!-- 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>
Dynamic novel view synthesis aims to capture the temporal evolution of visual content within videos.
Existing methods struggle to distinguishing between motion and structure, particularly in scenarios where camera poses are either unknown or constrained compared to object motion.
Furthermore, with information solely from reference images, it is extremely challenging to hallucinate unseen regions that are occluded or partially observed in the given videos.
To address these issues, we first finetune a pretrained RGB-D diffusion model on the video frames using a customization technique.
Subsequently, we distill the knowledge from the finetuned model to a 4D representations encompassing both dynamic and static Neural Radiance Fields (NeRF) components.
The proposed pipeline achieves geometric consistency while preserving the scene identity.
We perform thorough experiments to evaluate the efficacy of the proposed method qualitatively and quantitatively. Our results demonstrate the robustness and utility of our approach in challenging cases, further advancing dynamic novel view synthesis.
</p>
</div>
</div>
</div>
<!--/ Abstract. -->
</section>
<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">Method</h2>
<img src="./resources/images/framework.png">
<h2 class="content has-text-justified">
To perform dynamic novel view synthesis given a video, we adopt a 4D representation consisting of
dynamic and static parts. We use two types of supervision. First, we render the input viewpoints at input time. Besides, we distill prior
knowledge of a pre-trained RGB-D diffusion model on random novel views using score distillation sampling. Furthermore, to mitigate the
domain gaps between the training distributions and in-the-wild images, we tune the RGB-D diffusion model using the reference images
with a customization technique prior to distillation.
</h2>
</div>
</div>
</div>
<!-- </section> -->
<!-- <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">Results</h2> -->
<table align=center width=900px>
<!-- <center><h2 class="title is-3">View synthesis results on the iPhone dataset</h2></center> -->
<tr>
<center>
<div style='width:90%; text-align:justify; float:center'>
</center>
</tr>
<br>
<tr>
<th style='font-size: 22px' align="center" colspan="4">Stablized view</th>
<th style='font-size: 22px' align="center">Bullet-time</th>
</tr>
<tr>
<!-- <th style='font-size: 22px' align="center">Train View</th> -->
<th style='font-size: 22px' align="center">T-NeRF</th>
<th style='font-size: 22px' align="center">Nerfies</th>
<th style='font-size: 22px' align="center">HyperNeRF</th>
<th style='font-size: 22px' align="center">DpDy (Ours)</th>
<th style='font-size: 22px' align="center">DpDy (Ours)</th>
</tr>
<tr>
<!-- <td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-apple-fixed.mp4'>
</center>
</td> -->
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/tnerf-apple-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/nerfies-apple-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-apple-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-apple-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-apple.mp4'>
</center>
</td>
</tr>
<tr>
<!-- <td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-block-fixed.mp4'>
</center>
</td> -->
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/tnerf-block-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/nerfies-block-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-block-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-block-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-block.mp4'>
</center>
</td>
</tr>
<tr>
<!-- <td>
<center>
<video id="short_video" class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-paper-windmill-fixed.mp4'>
</center>
</td> -->
<td>
<center>
<video id="short_video" class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/tnerf-paper-windmill-fixed.mp4'>
</center>
</td>
<td>
<center>
<video id="short_video" class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/nerfies-paper-windmill-fixed.mp4'>
</center>
</td>
<td>
<center>
<video id="short_video" class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-paper-windmill-fixed.mp4'>
</center>
</td>
<td>
<center>
<video id="short_video" class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-paper-windmill-fixed.mp4'>
</center>
</td>
<td>
<center>
<video id="short_video" class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-paper-windmill.mp4'>
</center>
</td>
</tr>
<tr>
<!-- <td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-space-out-fixed.mp4'>
</center>
</td> -->
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/tnerf-space-out-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/nerfies-space-out-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-space-out-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-space-out-fixed.mp4'>
</center>
</td>
<td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-space-out.mp4'>
</center>
</td>
</tr>
<tr>
<!-- <td>
<center>
<video class='round' autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-teddy-fixed.mp4'>
</center>
</td> -->
<td>
<center>
<video autoplay muted loop playsinline style='width:150px' src='./resources/user-study/tnerf-teddy-fixed.mp4'>
</center>
</td>
<td>
<center>
<video autoplay muted loop playsinline style='width:150px' src='./resources/user-study/nerfies-teddy-fixed.mp4'>
</center>
</td>
<td>
<center>
<video autoplay muted loop playsinline style='width:150px' src='./resources/user-study/hypernerf-teddy-fixed.mp4'>
</center>
</td>
<td>
<center>
<video autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-teddy-fixed.mp4'>
</center>
</td>
<td>
<center>
<video autoplay muted loop playsinline style='width:150px' src='./resources/user-study/ours-teddy.mp4'>
</center>
</td>
</tr>
</table>
<!-- </div>
</div>
</div> -->
</section>
<footer class="footer">
<div class="container">
<div class="columns is-centered">
<div class="column is-8">
<div class="content">
<p>
This website is built using the template of <a
href="https://github.com/nerfies/nerfies.github.io">Nerfies</a>.
</p>
</div>
</div>
</div>
</div>
</footer>
<script>
// Get the video element
var video = document.getElementById("short_video");
// Set the loop attribute to ensure continuous looping
video.loop = true;
// Set the playback range between frames 0 and 100
var startFrame = 0;
var endFrame = 1;
// Listen for the "timeupdate" event to continuously check the playback position
video.addEventListener("timeupdate", function() {
// Check if the current time is beyond the specified range
if (video.currentTime > endFrame) {
// Set the playback position back to the start frame
video.currentTime = startFrame;
}
});
</script>
</body>
</html>