-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.html
More file actions
311 lines (270 loc) · 13.6 KB
/
index.html
File metadata and controls
311 lines (270 loc) · 13.6 KB
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
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="description" content="Human3D">
<meta name="keywords" content="Human3D">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>Human3D 🧑🤝🧑</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/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">
<link rel="stylesheet" href="css/style.css"> <!-- Resource style -->
<script src="js/modernizr.js"></script> <!-- Modernizr -->
<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>
<style>
.rcorners1 {
border-radius: 10px;
background: #ffffffd0;
padding: 5px;
font-size: 120%;
color: #5c5c5c;
}
</style>
</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">Human3D 🧑🤝🧑<br />
<p class="title is-3 publication-title">3D Segmentation of Humans in Point Clouds with Synthetic Data</p>
</h1>
<h1 class="title is-4" style="color: #5c5c5c;">ICCV 2023</h1>
<div class="is-size-5 publication-authors">
<span class="author-block">
<a href="https://aycatakmaz.github.io/">Ayça Takmaz</a><sup>*1</sup>,</span>
<span class="author-block">
<a href="https://jonasschult.github.io/">Jonas Schult</a><sup>*2</sup>,
</span>
<span class="author-block">
<a href="https://ikaftan.github.io/">Irem Kaftan</a><sup>†1</sup>,</span>
<span class="author-block">
<a href="https://cmakcay.github.io/">Mertcan Akçay</a><sup>†1</sup>,
</span>
<span class="author-block">
<a href="https://www.vision.rwth-aachen.de/person/1/">Bastian Leibe</a><sup>2</sup>,
</span>
<span class="author-block">
<a href="https://studios.disneyresearch.com/people/bob-sumner/">Robert Sumner</a><sup>1</sup>,
</span>
<br />
<span class="author-block">
<a href="https://francisengelmann.github.io/">Francis Engelmann</a><sup>1,3</sup>,
</span>
<span class="author-block">
<a href="https://inf.ethz.ch/people/person-detail.MjYyNzgw.TGlzdC8zMDQsLTg3NDc3NjI0MQ==.html">Siyu
Tang</a><sup>1</sup>
</span>
</div>
<div class="is-size-5 publication-authors">
<span class="author-block" style="margin-right: 1em;"><sup>1</sup>ETH Zürich, Switzerland</span>
<span class="author-block" style="margin-right: 1em;"><sup>2</sup>RWTH Aachen University, Germany</span>
<span class="author-block" style="margin-right: 1em;"><sup>1</sup>ETH AI Center, Switzerland</span>
<span class="author-block" style="margin-right: 1em;"><sup>*,†</sup>equal contribution</span>
</div>
<span class="link-block">
<a href="https://arxiv.org/abs/2212.00786" class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-file-pdf"></i>
</span>
<span>arXiv</span>
</a>
</span>
<span class="link-block">
<a href="assets/Human3D_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>
<span class="link-block">
<a href="#explanatory-video" class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fab fa-youtube"></i>
</span>
<span>Video</span>
</a>
</span>
<span class="link-block">
<a href="https://github.com/human-3d/Human3D" 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>
<span class="link-block">
<a href="https://human-3d.github.io/dataset" class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-database"></i>
</span>
<span>Dataset</span>
</a>
</span>
<span class="link-block">
<a href="https://omnomnom.vision.rwth-aachen.de/data/human3d/visualizations/egobody/"
class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fa fa-images"></i>
</span>
<span>EgoBody Visualizations</span>
<span class="link-block">
<a href="#bibtex" class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<i class="fas fa-book"></i>
</span>
<span>BibTeX</span>
</a>
</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="assets/teaser.jpg" style="max-width:100%" />
<h2 class="subtitle has-text-centered">
</h2>
</div>
</div>
</section>
<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>
<b>
TL;DR: We propose the first multi-human body-part segmentation model, called Human3D 🧑🤝🧑, that
directly operates on 3D scenes.
In an extensive analysis, we validate the benefits of training on synthetic data on multiple baselines
and tasks.
</b>
<br /><br />
Segmenting humans in 3D indoor scenes has become increasingly important with the rise of human-centered robotics and AR/VR applications. To this end, we propose the task of joint 3D human semantic segmentation, instance segmentation and multi-human body-part segmentation. Few works have attempted to directly segment humans in cluttered 3D scenes, which is largely due to the lack of annotated training data of humans interacting with 3D scenes. We address this challenge and propose a framework for generating training data of synthetic humans interacting with real 3D scenes. Furthermore, we propose a novel transformer-based model, Human3D, which is the first end-to-end model for segmenting multiple human instances and their body-parts in a unified manner. The key advantage of our synthetic data generation framework is its ability to generate diverse and realistic human-scene interactions, with highly accurate ground truth. Our experiments show that pre-training on synthetic data improves performance on a wide variety of 3D human segmentation tasks. Finally, we demonstrate that Human3D outperforms even task-specific state-of-the-art 3D segmentation methods.
</p>
</div>
</div>
</div>
<!--/ Abstract. -->
<!-- Concurrent Work. -->
<div class="columns is-centered">
<div class="column is-full-width">
<p id="explanatory-video"></p>
<h2 class="title is-3">Explanatory Video</h2>
<video controls poster="assets/explanatory_video_poster.jpg">
<source src="assets/Human3D_explanatory_video.mp4" type="video/mp4">
Your browser does not support the video tag.
</video>
<h2 class="title is-4">Point Cloud from iPhone LiDAR</h2>
<video autoplay muted loop>
<source src="assets/poster_session.mp4" type="video/mp4">
Your browser does not support the video tag.
</video>
<div class="content has-text-justified">
<p>
Remarkably, our approach generalizes to out-of-distribution examples. Although trained on synthetic data
and real Kinect depth data, Human3D shows promising results on reconstructed point clouds scanned with an
iPhone LiDAR sensor.
</p>
</div>
<h2 class="title is-4">Depth from Kinect Sensor</h2>
<video autoplay muted loop>
<source src="assets/kinect_video.mp4" type="video/mp4">
Your browser does not support the video tag.
</video>
<div class="content has-text-justified">
<p>
Human3D shows smooth and robust predictions on videos recorded with the Kinect Depth Sensor.
</p>
</div>
<h2 class="title is-4">Synthetic Pretraining</h2>
<figure class="cd-image-container">
<img src="assets/without_synthetic_pretraining.jpg" alt="Original Image">
<span class="cd-image-label" data-type="original">
</span>
<div class="cd-resize-img" style="border-right: 2px dotted rgb(94, 94, 94);">
<!-- the resizable image on top -->
<img src="assets/with_synthetic_pretraining.jpg" alt="Modified Image">
<span class="cd-image-label" data-type="modified">
</span>
</div>
<span class="cd-handle"></span>
</figure> <!-- cd-image-container -->
<div class="content has-text-justified">
<p>
<b>Only EgoBody data:</b>
We observe that models trained only on EgoBody data do not generalize to scenes with more than 2 humans. Here we can
see that the instance masks of two people leak into the third person's mask on the right.
The reason for this is that the EgoBody dataset only contains scenes with less than 3 people at the same time. When
only trained on EgoBody, Human3D inevitably learns this bias and consequently fails for scenes with more than 2
people.
<br />
<b>Pretrained with synthetic data:</b>
In contrast, our synthetic dataset consists of scenes with up to 10 people. Human3D, pre-trained on synthetic data
and fine-tuned on real EgoBody data, shows significantly better results for scenes with a larger number of people.
<br /> <br />
<b>We conclude that pre-training with synthetic data helps to segment humans in 3D point clouds!</b>
</p>
</div>
<h2 class="title is-3">Related Work</h2>
<div class="content has-text-justified">
<a href="https://caizhongang.github.io/projects/PointHPS/">PointHPS: Cascaded 3D Human Pose and Shape Estimation from Point Clouds</a>
</div>
<h2 class="title is-3">Publication</h2>
<div class="content has-text-justified">
<a href="https://arxiv.org/abs/2212.00786"><img src="assets/paper_preview.jpg" style="max-width:100%" /></a>
</div>
<p id="bibtex"></p>
<h2 class="title is-3">BibTeX</h2>
<pre><code>
@inproceedings{human3d,
title = {{3D Segmentation of Humans in Point Clouds with Synthetic Data}},
author = {Takmaz, Ay\c{c}a and Schult, Jonas and Kaftan, Irem and Ak\c{c}ay, Mertcan
and Leibe, Bastian and Sumner, Robert and Engelmann, Francis and Tang, Siyu},
booktitle = {{Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)}},
year = {2023}
}
</code></pre>
</div>
</div>
<!--/ Concurrent Work. -->
<footer class="footer">
<div class="container">
<div class="columns is-centered">
<div class="column is-8">
<div class="content">
<p>
This website 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>
It borrows the source code of <a href="https://github.com/nerfies/nerfies.github.io">this website</a>.
We would like to thank Utkarsh Sinha and Keunhong Park.
</p>
</div>
</div>
</div>
</div>
</footer>
</body>
<script src="js/jquery-2.1.1.js"></script>
<script src="js/jquery.mobile.custom.min.js"></script> <!-- Resource jQuery -->
<script src="js/main.js"></script> <!-- Resource jQuery -->
</html>