-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathindex.html
More file actions
executable file
·1032 lines (970 loc) · 45.4 KB
/
index.html
File metadata and controls
executable file
·1032 lines (970 loc) · 45.4 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
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
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
<!DOCTYPE html>
<html>
<head>
<meta charset="utf-8">
<meta name="description"
content="REACT3D: Recovering Articulations for Interactive Physical 3D Scenes">
<meta name="keywords" content="REACT3D, interactive 3D scene, articulation estimation, part segmentation, point cloud segmentation, functionality understanding, affordance understanding, functional interactive elements">
<meta name="viewport" content="width=device-width, initial-scale=1">
<title>REACT3D</title>
<!-- Google tag (gtag.js) -->
<script async src="https://www.googletagmanager.com/gtag/js?id=G-RS1JGNX8ND"></script>
<script>
window.dataLayer = window.dataLayer || [];
function gtag(){dataLayer.push(arguments);}
gtag('js', new Date());
gtag('config', 'G-RS1JGNX8ND');
</script>
<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"> -->
<link rel="apple-touch-icon" sizes="180x180" href="static/images/logo.png">
<link rel="icon" type="image/png" sizes="32x32" href="static/images/logo.png">
<link rel="icon" type="image/png" sizes="16x16" href="static/images/logo.png">
<!-- <link rel="manifest" href="./static/images/favicon_io/site.webmanifest"> -->
<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>
<style>
html {
scroll-behavior: smooth;
}
</style>
</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://keunhong.com">
<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://hypernerf.github.io">
HyperNeRF
</a>
<a class="navbar-item" href="https://nerfies.github.io">
Nerfies
</a>
<a class="navbar-item" href="https://latentfusion.github.io">
LatentFusion
</a>
<a class="navbar-item" href="https://photoshape.github.io">
PhotoShape
</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">REACT3D</h1> -->
<h1 class="title is-1 publication-title" style="text-align: center; display: flex; justify-content: center; align-items: center; flex-wrap: wrap;"><span style="margin-top: 7px">REACT3D</span><img src="static/images/logo.png" alt="Logo" width="80px" style="margin-left: 20px;"></h1>
<h1 class="title is-3 publication-title">Recovering Articulations for Interactive Physical 3D Scenes</h1>
<h1 class="title is-4" style="color: #5c5c5c;">Accepted at IEEE Robotics and Automation Letters (RA-L)</h1>
<div class="is-size-5 publication-authors">
<span class="author-block">
<a href="https://troyehuang.github.io/">Zhao Huang</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="https://boysun045.github.io/boysun-website/">Boyang Sun</a><sup>1</sup>,
</span>
<br>
<span class="author-block">
<a href="https://alexdelitzas.github.io">Alexandros Delitzas</a><sup>1,2</sup>,
</span>
<span class="author-block">
<a href="https://www.linkedin.com/in/chen-jiaqi/">Jiaqi Chen</a><sup>1</sup>,
</span>
<span class="author-block">
<a href="https://people.inf.ethz.ch/pomarc/">Marc Pollefeys</a><sup>1,3</sup>
</span>
</div>
<div class="is-size-5 publication-authors">
<span class="author-block" style="margin-right: 1em;"><sup>1</sup>ETH Zürich</span>
<span class="author-block" style="margin-right: 1em;"><sup>2</sup>Max Planck Institute for Informatics</span>
<span class="author-block" style="margin-right: 1em;"><sup>3</sup>Microsoft</span>
</div>
<div class="column has-text-centered">
<div class="publication-links">
<!-- <span class="link-block">
<a href=""
class="external-link button is-normal is-rounded is-dark" disabled="">
<span class="icon">
<i class="ai ai-arxiv"></i>
</span>
<span>arXiv</span>
</a>
</span> -->
<!-- Video Link. -->
<!-- <span class="link-block">
<a href="https://www.youtube.com/watch?v=MrKrnHhk8IA"
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://alexdelitzas.github.io/assets/pdf/SceneFun3D.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 class="external-link button is-normal is-rounded is-dark" href="https://ieeexplore.ieee.org/document/11434845" target="_blank">
<!-- https://arxiv.org/abs/2510.11340 -->
<span class="icon"><i class="fas fa-file-pdf"></i></span><span>Paper</span>
</a>
</span>
<span class="link-block">
<a class="external-link button is-normal is-rounded is-dark" href="https://arxiv.org/abs/2510.11340" target="_blank">
<span class="icon"><i class="ai ai-arxiv"></i></span><span>arXiv</span>
</a>
</span>
<span class="link-block">
<a href="#simulator-videos" 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="#interactive-demo" class="external-link button is-normal is-rounded is-dark">
<span class="icon"><i class="fas fa-cube"></i></span><span>3D Demo</span>
</a>
</span>
<span class="link-block">
<a href="https://github.com/troyehuang/REACT3D" 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="./static/images/poster.pdf" class="external-link button is-normal is-rounded is-dark">
<span class="icon">
<svg class="svg-inline--fa fa-palette fa-w-16" aria-hidden="true" focusable="false" data-prefix="fas" data-icon="palette" role="img" xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" data-fa-i2svg=""><path fill="currentColor" d="M204.3 5C104.9 24.4 24.8 104.3 5.2 203.4c-37 187 131.7 326.4 258.8 306.7 41.2-6.4 61.4-54.6 42.5-91.7-23.1-45.4 9.9-98.4 60.9-98.4h79.7c35.8 0 64.8-29.6 64.9-65.3C511.5 97.1 368.1-26.9 204.3 5zM96 320c-17.7 0-32-14.3-32-32s14.3-32 32-32 32 14.3 32 32-14.3 32-32 32zm32-128c-17.7 0-32-14.3-32-32s14.3-32 32-32 32 14.3 32 32-14.3 32-32 32zm128-64c-17.7 0-32-14.3-32-32s14.3-32 32-32 32 14.3 32 32-14.3 32-32 32zm128 64c-17.7 0-32-14.3-32-32s14.3-32 32-32 32 14.3 32 32-14.3 32-32 32z"></path></svg>
</span><span>Poster</span>
</a>
</span> -->
<!-- <span class="link-block">
<a href="https://scenefun3d.github.io/documentation" class="external-link button is-normal is-rounded is-dark">
<span class="icon"><i class="fa fa-book" aria-hidden="true"></i></span><span>Dataset</span>
</a>
</span> -->
<!-- <span class="link-block">
<a href="https://scenefun3d.github.io/benchmark" class="external-link button is-normal is-rounded is-dark">
<span class="icon"><i class="fa fa-chart-line" aria-hidden="true"></i></span><span>Benchmark</span>
</a>
</span> -->
<!-- <span class="link-block">
<a href="https://eval.ai/web/challenges/challenge-page/2466" class="external-link button is-normal is-rounded is-dark">
<span class="icon"><i class="fa fa-rocket" aria-hidden="true"></i></span><span>EvalAI</span>
</a>
</span> -->
</div>
</div>
</div>
</div>
</div>
<br />
<br />
<div class="container is-max-desktop">
<video width="100%" autoplay muted loop playsinline controls>
<source src="./recordings/promotion.mp4" type="video/mp4">
</video>
<br /><br /><br />
<h2 class="subtitle has-text-centered">
We introduce <strong>REACT3D</strong>, a scalable zero-shot framework that <br/> converts static 3D scenes into <strong>simulation-ready interactive replicas</strong> with consistent geometry.
</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-steve">
<video poster="" id="steve" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/steve.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-chair-tp">
<video poster="" id="chair-tp" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/chair-tp.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-shiba">
<video poster="" id="shiba" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/shiba.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-fullbody">
<video poster="" id="fullbody" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/fullbody.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-blueshirt">
<video poster="" id="blueshirt" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/blueshirt.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-mask">
<video poster="" id="mask" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/mask.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-coffee">
<video poster="" id="coffee" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/coffee.mp4"
type="video/mp4">
</video>
</div>
<div class="item item-toby">
<video poster="" id="toby" autoplay controls muted loop playsinline height="100%">
<source src="./static/videos/toby2.mp4"
type="video/mp4">
</video>
</div>
</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">Abstract</h2>
<div class="content has-text-justified">
<img src="./static/images/teaser.png" class="teaser-fig" alt="teaser-fig." />
<!-- <img src="./static/images/teaser_v7_3.png" class="teaser-fig" alt="teaser-fig." /> -->
<div style="margin: -80px 230px 100px 230px">
</div>
<p>
Interactive 3D scenes are increasingly vital for embodied intelligence, yet existing datasets remain limited due to the labor-intensive process of annotating part segmentation, kinematic types, and motion trajectories. We present REACT3D, a scalable zero-shot framework that converts static 3D scenes into simulation-ready interactive replicas with consistent geometry, enabling direct use in diverse downstream tasks. Our contributions include: (i) openable-object detection and segmentation to extract candidate movable parts from static scenes, (ii) articulation estimation that infers joint types and motion parameters, (iii) hidden-geometry completion followed by interactive object assembly, and (iv) interactive scene integration in widely supported formats to ensure compatibility with standard simulation platforms. We achieve state-of-the-art performance on detection/segmentation and articulation metrics across diverse indoor scenes, demonstrating the effectiveness of our framework and providing a practical foundation for scalable interactive scene generation, thereby lowering the barrier to large-scale research on articulated scene understanding.
</p>
</div>
</div>
</div>
</div>
</section>
<!-- <section class="section">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column">
<h2 class="title is-3">Video</h2>
<div class="publication-video">
<iframe src="https://www.youtube.com/embed/T-HPNIYUOoc" title="[CVPR '24 Oral] Fine-Grained Functionality and Affordance Understanding in 3D Scenes" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share" referrerpolicy="strict-origin-when-cross-origin" allowfullscreen></iframe>
</div>
</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"> -->
<div class="column">
<hr />
<h2 class="title is-3">Motivation</h2>
<!-- <div class="columns is-centered">
<div class="column">
<div class="content">
<video autoplay muted loop poster="static/images/robot_drawer_pic.png">
<source src="static/images/robot_drawer_motivation.mp4" type="video/mp4">
</video>
</div>
</div>
<div class="column">
<div class="content">
<video autoplay muted loop poster="static/images/robot_floorlamp_pic.png">
<source src="static/images/robot_floorlamp_motivation.mp4" type="video/mp4">
</video>
</div>
</div>
</div> -->
<div class="content has-text-justified">
<p>
Embodied AI requires <strong>large-scale, physically interactive 3D environments</strong>, but building them today hinges on <strong>slow, expert-heavy annotations</strong> of parts, joints, and motions. Meanwhile, <strong>vast static 3D datasets</strong> already exist yet remain inert for interaction and simulation. Our motivation is to <strong>unlock the latent interactivity</strong> of these static assets—recovering articulations and consistent geometry automatically—so that researchers can train and evaluate agents at scale <strong>without new data collection or manual labeling</strong>. By converting abundant static scenes into reliable interactive environments, we aim to <strong>collapse the data bottleneck</strong> that currently limits progress in articulated scene understanding and embodied intelligence.
</p>
</div>
</div>
</div>
</div>
</section>
<section class="section">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column">
<h2 class="title is-3">REACT3D</h2>
<!-- <img src="./static/images/framework.png" class="teaser-fig" alt="teaser-fig."> -->
<div class="content has-text-justified">
<p>
We propose a novel, application-driven framework for converting high-quality static scenes into physics-aware, interactive environments:
<ul>
<li><strong>Methodology</strong>: REACT3D detects openable objects, segments movable parts, infers joints and motion, completes hidden geometry, and exports simulation-ready assets.</li>
<li><strong>Performance</strong>: We achieve state-of-the-art detection/segmentation and articulation metrics across diverse indoor scenes, against strong baselines.</li>
<li><strong>Utility</strong>: The resulting assets integrate with ROS, Isaac Sim, PyBullet, and Open3D, enabling large-scale training and evaluation without new data or manual labels.</li>
</ul>
</p>
</div>
<br />
<h3 class="title is-4">System Overview</h3>
<!-- <div class="container is-max-desktop"> -->
<img src="./static/images/REACT3D_overview.png" class="teaser-fig is-max-desktop" alt="teaser-fig.">
<!-- </div> -->
<br /><br />
<div class="content has-text-justified">
<p>
Given a static 3D scene, our method first applies open-vocabulary detection to identify openable objects and segmentation to extract their movable parts. We then estimate articulations and generate hidden geometry to obtain interactive objects. Finally, they are integrated with the static background to produce a simulation-ready interactive scene.
</p>
</div>
<br />
<h3 class="title is-4">Interactive Object Generation</h3>
<div class="columns is-mobile is-multiline is-centered" style="align-items: flex-end;">
<div class="column is-one-sixth">
<p class="has-text-centered mb-5" style="font-size: 16px;"><strong>Predicted 2D Mask</strong></p>
<img src="./static/images/2d_to_open/image_to_open_image.png" alt="Step 1" style="height:125px; object-fit:contain; width:auto; max-width: none !important;">
</div>
<div class="column is-one-sixth">
<p class="has-text-centered" style="font-size: 16px;"><strong>Segmented Mesh</strong></p>
<img src="./static/images/2d_to_open/image_to_open_mesh.png" alt="Step 2" style="height:150px; object-fit:contain; width:auto;">
</div>
<div class="column is-one-sixth">
<p class="has-text-centered" style="font-size: 16px;"><strong>Predicted Joint</strong></p>
<img src="./static/images/2d_to_open/image_to_open_opdm.png" alt="Step 3" style="height:150px; object-fit:contain; width:auto;">
</div>
<div class="column is-one-sixth">
<p class="has-text-centered" style="font-size: 16px;"><strong>Refined Joint</strong></p>
<img src="./static/images/2d_to_open/image_to_open_refined.png" alt="Step 4" style="height:150px; object-fit:contain; width:auto;">
</div>
<div class="column is-one-sixth">
<p class="has-text-centered" style="font-size: 16px;"><strong>Interactive Object</strong></p>
<img src="./static/images/2d_to_open/image_to_open_interactive.png" alt="Step 5" style="height:150px; object-fit:contain; width:auto;">
</div>
<div class="column is-one-sixth">
<p class="has-text-centered" style="font-size: 16px;"><strong>Opened Object</strong></p>
<img src="./static/images/2d_to_open/image_to_open_open_2.png" alt="Step 6" style="height:150px; object-fit:contain; width:auto;">
</div>
</div>
<br /><br />
<div class="content has-text-justified">
<p>
From left to right, the figure shows key intermediate results of interactive object generation. In the last column, the thin red line highlights the contour of the base part.
</p>
</div>
<br />
<h3 class="title is-4">Qualitative Results</h3>
<style>
.tight-columns {
row-gap: 0 !important;
}
.tight-columns .column:nth-child(n+4) {
margin-top: -20px !important;
}
</style>
<div class="columns is-mobile is-multiline is-centered tight-columns" style="align-items: flex-end;">
<div class="column is-one-third">
<!-- <p class="has-text-centered mb-5" style="font-size: 16px;"><strong>Predicted 2D Mask</strong></p> -->
<img src="./static/images/input_result_comparison/c0f5742640_scannetpp.jpg" alt="Input 1" style="width: 100%; height: auto; object-fit: contain;">
</div>
<div class="column is-one-third">
<!-- <p class="has-text-centered" style="font-size: 16px;"><strong>Segmented Mesh</strong></p> -->
<img src="./static/images/input_result_comparison/3c95c89d61_scannetpp_2.jpg" alt="Input 2" style="width: 100%; height: auto; object-fit: contain;">
</div>
<div class="column is-one-third">
<!-- <p class="has-text-centered" style="font-size: 16px;"><strong>Predicted Joint</strong></p> -->
<img src="./static/images/input_result_comparison/7f4d173c9c_scannetpp.jpg" alt="Input 3" style="width: 100%; height: auto; object-fit: contain;">
</div>
<!-- <br /> -->
<div class="column is-one-third">
<!-- <p class="has-text-centered" style="font-size: 16px;"><strong>Refined Joint</strong></p> -->
<img src="./static/images/input_result_comparison/c0f5742640_ours_isaac.jpg" alt="Output 1" style="width: 100%; height: auto; object-fit: contain;">
</div>
<div class="column is-one-third">
<!-- <p class="has-text-centered" style="font-size: 16px;"><strong>Interactive Object</strong></p> -->
<img src="./static/images/input_result_comparison/3c_ours_isaac.jpg" alt="Output 2" style="width: 100%; height: auto; object-fit: contain;">
</div>
<div class="column is-one-third">
<!-- <p class="has-text-centered" style="font-size: 16px;"><strong>Opened Object</strong></p> -->
<img src="./static/images/input_result_comparison/7f4d173c9c_ours_isaac.jpg" alt="Output 3" style="width: 100%; height: auto; object-fit: contain;">
</div>
</div>
<div class="content has-text-justified">
<p>
We show three example input scenes from ScanNet++ (top row) and their corresponding interactive replicas generated by REACT3D (bottom row), visualized in Isaac Sim. Our method robustly converts diverse object types, shapes, sizes, and articulation mechanisms, producing high-quality interactive scenes.
</p>
</div>
</div>
</div>
</div>
</section>
<section class="section" id="simulator-videos">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column">
<h2 class="title is-3">Manipulating Interactive Scenes in Simulator</h2>
<!-- <img src="./static/images/framework.png" class="teaser-fig" alt="teaser-fig."> -->
<div class="content has-text-justified">
<p>
We showcase the interactive scenes generated by REACT3D in two widely used simulators, Isaac Sim and ROS. The videos demonstrate the physical interactivity of various objects. Users can manipulate each single object through GUI, highlighting the versatility and applicability of our generated scenes in different simulation environments.
</p>
</div>
<br />
<style>
.video-container {
width: 100%;
max-width: 960px;
margin: 0 auto;
}
.video-container video {
width: 100%;
height: auto;
aspect-ratio: 960 / 505;
object-fit: cover;
}
</style>
<h3 class="title is-4">Isaac Sim</h3>
<div class="video-container">
<video id="isaac_v1" controls loop>
<source src="static/videos/3c_isaacsim.mp4" type="video/mp4">
</video>
<video id="isaac_v2" controls loop hidden>
<source src="static/videos/7f_isaacsim.mp4" type="video/mp4">
</video>
<video id="isaac_v3" controls loop hidden>
<source src="static/videos/c0f_isaacsim.mp4" type="video/mp4">
</video>
</div>
<div class="buttons is-centered mt-3">
<button class="button is-dark is-squared video-changer" onclick="show('isaac_v1','isaac')">
Video 1
</button>
<button class="button is-dark is-squared video-changer" onclick="show('isaac_v2','isaac')">
Video 2
</button>
<button class="button is-dark is-squared video-changer" onclick="show('isaac_v3','isaac')">
Video 3
</button>
</div>
<br />
<h3 class="title is-4 ">ROS</h3>
<div class="video-container">
<video id="ros_v1" controls loop>
<source src="static/videos/ROS_GUI_demo_3c95c89d61.webm" type="video/webm">
</video>
<video id="ros_v2" controls loop hidden>
<source src="static/videos/ROS_GUI_demo_7f4d173c9c.webm" type="video/webm">
</video>
<video id="ros_v3" controls loop hidden>
<source src="static/videos/ROS_GUI_demo_40aec5fffa.webm" type="video/webm">
</video>
</div>
<div class="buttons is-centered mt-3">
<button class="button is-dark is-squared video-changer" onclick="show('ros_v1','ros')">
Video 1
</button>
<button class="button is-dark is-squared video-changer" onclick="show('ros_v2','ros')">
Video 2
</button>
<button class="button is-dark is-squared video-changer" onclick="show('ros_v3','ros')">
Video 3
</button>
</div>
</div>
</div>
</div>
</section>
<section class="section" id="interactive-demo">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column">
<h2 class="title is-3">Interactive 3D Demo</h2>
<div class="content has-text-justified">
<p>
Explore the generated interactive scene directly in your browser. You can rotate <span class="icon is-small"><i class="fas fa-sync"></i></span>, zoom <span class="icon is-small"><i class="fas fa-search-plus"></i></span>, and pan <span class="icon is-small"><i class="fas fa-arrows-alt"></i></span> to observe the interactive scene generated by <strong>REACT3D</strong>.
</p>
</div>
<div style="position: relative; padding-bottom: 56.25%; height: 0; overflow: hidden; max-width: 100%; border-radius: 12px; box-shadow: 0 4px 12px rgba(0,0,0,0.15); border: 1px solid #eee;">
<iframe
src="https://react3d.github.io/viser-client/?playbackPath=https://react3d.github.io/recordings/recording.viser&initialCameraPosition=2.589,-2.683,1.833&initialCameraLookAt=2.634,1.154,1.556&initialCameraUp=0.000,0.000,1.000&initialCameraFov=1.3090&initialCameraNear=0.01&initialCameraFar=1000"
style="position: absolute; top: 0; left: 0; width: 100%; height: 100%; border: none;"
allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture"
allowfullscreen>
</iframe>
</div>
<p class="is-size-6 mt-4">
This interactive demo is supported by <a href="https://viser.studio/main/" target="_blank">Viser</a>.
</p>
</div>
</div>
</div>
</section>
<!-- <section class="section">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column">
<h2 class="title is-3">Applications</h2>
<img src="./static/images/potential_applications.png" class="teaser-fig" alt="teaser-fig.">
<div class="content has-text-justified">
<p>
Our aim is to <strong>catalyze research</strong> on robotic systems, embodied AI and AR/VR applications.
In robotics (left), localizing visual affordances and grounding them to natural language task
descriptions is a crucial skill for embodied intelligent agents. By providing accurate 3D affordance masks,
we facilitate the generation of realistic human-scene interactions (right), unleashing new directions in virtual human applications.
</p>
</div>
</div>
</div>
</div>
</section> -->
<section class="section" id="Quantitative-Results">
<div class="container is-max-desktop">
<div class="columns is-centered has-text-centered">
<div class="column">
<h2 class="title is-3">Quantitative Results</h2>
<div class="content has-text-justified">
<p>
<b>Quantitative results on openable object detection on ScanNet++.</b> Openable object detection on 30 ScanNet++ scenes rich in openable objects. We report Precision, Recall, and F1 at IoU thresholds τ∈{0.25, 0.50}. <sup>*</sup>For URDFormer, τ denotes the coverage ratio, while for other methods it denotes IoU. <sup>†</sup>For URDFormer, the number reflects only openable objects visible in the selected region with the highest density of openable objects, since the method takes a single RGB image as input. <sup>‡</sup>denotes the use of ground-truth mesh as input and the Gemini 2.5 model for detection. <span class="best">Best</span>, <span class="second">second-best</span>, and <span class="third">third-best</span> results are highlighted.
</p>
</div>
<div class="table-container mb-6">
<table class="academic-table table is-fullwidth">
<thead>
<tr>
<th rowspan="2" style="vertical-align: middle; text-align: left;">Method</th>
<th rowspan="2" style="vertical-align: middle;">#Openable<br>Objects</th>
<th colspan="3" style="border-bottom: 1px solid black;">τ<sup>*</sup>=0.25</th>
<th colspan="3" style="border-bottom: 1px solid black;">τ<sup>*</sup>=0.50</th>
</tr>
<tr>
<th>Precision↑</th>
<th>Recall↑</th>
<th>F1 Score↑</th>
<th>Precision↑</th>
<th>Recall↑</th>
<th>F1 Score↑</th>
</tr>
</thead>
<tbody>
<tr class="midrule">
<td style="text-align: left;">URDFormer</td>
<td>226<sup>†</sup></td>
<td class="third">0.429</td>
<td class="second">0.226</td>
<td class="second">0.296</td>
<td class="second">0.361</td>
<td class="second">0.190</td>
<td class="second">0.249</td>
</tr>
<tr>
<td style="text-align: left;">DRAWER (GT-mesh 2.5<sup>‡</sup>)</td>
<td>540</td>
<td class="second">0.434</td>
<td class="third">0.141</td>
<td class="third">0.213</td>
<td class="third">0.326</td>
<td class="third">0.106</td>
<td class="third">0.159</td>
</tr>
<tr>
<td style="text-align: left;">REACT3D (Ours)</td>
<td>540</td>
<td class="best">0.731</td>
<td class="best">0.393</td>
<td class="best">0.511</td>
<td class="best">0.666</td>
<td class="best">0.357</td>
<td class="best">0.465</td>
</tr>
</tbody>
</table>
</div>
<div class="content has-text-justified">
<p>
<b>Quantitative results of articulation estimation and MOD evaluation on ScanNet++.</b> Articulation evaluation on the same 30 scenes from ScanNet++. Joint type accuracy is computed over correctly detected movable parts (+M / P<sub>Det</sub>), while minimum distance (MD) and orientation error (OE) are reported only for parts with correctly predicted joint types. Results are broken down by joint types: Prismatic, Revolute Vertical (Vert.), and Revolute Horizontal (Hor.). <sup>*</sup><sup>†</sup><sup>‡</sup>Conditions are the same as in our initial evaluation. <span class="best">Best</span>, <span class="second">second-best</span>, and <span class="third">third-best</span> results are highlighted.
</p>
</div>
<style>
.table-container {
overflow-x: auto;
}
.academic-table {
width: 100%;
border-collapse: collapse;
margin: 0 auto;
font-size: 0.9em;
}
.academic-table th, .academic-table td {
padding: 8px 4px;
text-align: center;
border: none;
vertical-align: middle;
white-space: nowrap;
}
.academic-table th {
font-weight: bold;
}
.academic-table thead tr:first-child th {
border-top: 2px solid black;
}
.academic-table thead tr:last-child th {
border-bottom: 1px solid black;
}
.academic-table tbody tr:last-child td {
border-bottom: 2px solid black;
}
.academic-table tbody tr.midrule td {
border-top: 1px solid black;
}
.best { font-weight: bold; }
.second { text-decoration: underline; }
.third { font-style: italic; }
</style>
<div class="table-container">
<table class="academic-table table is-fullwidth">
<thead>
<tr>
<th rowspan="3" style="vertical-align: middle; text-align: left;">Method</th>
<th rowspan="3" style="vertical-align: middle;">#Openable<br>Objects</th>
<th colspan="7" style="border-bottom: 1px solid black;">τ<sup>*</sup>=0.25</th>
<th colspan="7" style="border-bottom: 1px solid black;">τ<sup>*</sup>=0.50</th>
</tr>
<tr>
<th colspan="3" style="border-bottom: 1px solid black;">Articulation</th>
<th colspan="4" style="border-bottom: 1px solid black;">MOD [%]↑</th>
<th colspan="3" style="border-bottom: 1px solid black;">Articulation</th>
<th colspan="4" style="border-bottom: 1px solid black;">MOD [%]↑</th>
</tr>
<tr>
<th>Joint Acc. [%]↑</th>
<th>MD [m]↓</th>
<th>OE [°]↓</th>
<th>PDet</th>
<th>+M</th>
<th>+MO</th>
<th>+MOD</th>
<th>Joint Acc. [%]↑</th>
<th>MD [m]↓</th>
<th>OE [°]↓</th>
<th>PDet</th>
<th>+M</th>
<th>+MO</th>
<th>+MOD</th>
</tr>
</thead>
<tbody>
<!-- URDFormer -->
<tr class="midrule">
<td style="text-align: left;"><b>URDFormer</b></td>
<td>226<sup>†</sup></td>
<td class="third">51.0</td>
<td class="third">0.583</td>
<td class="best">0.50</td>
<td class="second">22.6</td>
<td class="second">11.5</td>
<td class="second">11.5</td>
<td class="third">2.2</td>
<td class="third">46.5</td>
<td class="third">0.491</td>
<td class="best">0.61</td>
<td class="second">19.0</td>
<td class="second">8.8</td>
<td class="second">8.8</td>
<td class="third">3.1</td>
</tr>
<tr>
<td style="text-align: left; padding-left: 2em;"><i>Prismatic</i></td>
<td>63</td>
<td>--</td>
<td>--</td>
<td>2.05</td>
<td>9.5</td>
<td>1.6</td>
<td>1.6</td>
<td>1.6</td>
<td>--</td>
<td>--</td>
<td>1.95</td>
<td>12.7</td>
<td>1.6</td>
<td>1.6</td>
<td>1.6</td>
</tr>
<tr>
<td style="text-align: left; padding-left: 2em;"><i>Revolute (Vert.)</i></td>
<td>159</td>
<td>--</td>
<td>0.548</td>
<td>0.34</td>
<td>25.8</td>
<td>14.5</td>
<td>14.5</td>
<td>3.1</td>
<td>--</td>
<td>0.493</td>
<td>0.40</td>
<td>20.1</td>
<td>11.3</td>
<td>11.3</td>
<td>4.4</td>
</tr>
<tr>
<td style="text-align: left; padding-left: 2em;"><i>Revolute (Hor.)</i></td>
<td>4</td>
<td>--</td>
<td>0.452</td>
<td>3.06</td>
<td>50.0</td>
<td>25.0</td>
<td>25.0</td>
<td>0.0</td>
<td>--</td>
<td>0.452</td>
<td>3.06</td>
<td>50.0</td>
<td>25.0</td>
<td>25.0</td>
<td>0.0</td>
</tr>
<!-- DRAWER -->
<tr class="midrule">
<td style="text-align: left;"><b>DRAWER</b> (GT-mesh 2.5<sup>‡</sup>)</td>
<td>540</td>
<td class="second">73.7</td>
<td class="second">0.288</td>
<td class="third">1.62</td>
<td class="third">14.1</td>
<td class="third">10.4</td>
<td class="third">10.0</td>
<td class="second">5.2</td>
<td class="second">78.9</td>
<td>0.313</td>
<td class="third">1.71</td>
<td class="third">10.6</td>
<td class="third">8.3</td>
<td class="third">8.0</td>
<td class="second">3.7</td>
</tr>
<tr>
<td style="text-align: left; padding-left: 2em;"><i>Prismatic</i></td>
<td>136</td>
<td>--</td>
<td>--</td>
<td>--</td>
<td>12.5</td>
<td>0.0</td>
<td>0.0</td>
<td>0.0</td>
<td>--</td>
<td>--</td>
<td>--</td>
<td>7.4</td>
<td>0.0</td>
<td>0.0</td>
<td>0.0</td>
</tr>
<tr>
<td style="text-align: left; padding-left: 2em;"><i>Revolute (Vert.)</i></td>
<td>386</td>
<td>--</td>
<td>0.288</td>
<td>1.62</td>
<td>14.5</td>
<td>14.5</td>
<td>14.0</td>
<td>7.3</td>
<td>--</td>
<td>0.313</td>
<td>1.71</td>
<td>11.7</td>
<td>11.7</td>
<td>11.1</td>
<td>5.2</td>
</tr>
<tr>
<td style="text-align: left; padding-left: 2em;"><i>Revolute (Hor.)</i></td>
<td>18</td>
<td>--</td>
<td>--</td>
<td>--</td>
<td>16.7</td>
<td>0.0</td>
<td>0.0</td>
<td>0.0</td>
<td>--</td>
<td>--</td>
<td>--</td>
<td>11.1</td>
<td>0.0</td>
<td>0.0</td>
<td>0.0</td>
</tr>
<!-- REACT3D -->
<tr class="midrule">
<td style="text-align: left;"><b>REACT3D (Ours)</b></td>
<td>540</td>
<td class="best">91.5</td>
<td class="best">0.203</td>
<td class="second">1.16</td>
<td class="best">39.3</td>
<td class="best">35.9</td>
<td class="best">35.9</td>
<td class="best">25.2</td>
<td class="best">92.2</td>
<td class="best">0.180</td>
<td class="second">1.13</td>
<td class="best">35.7</td>
<td class="best">33.0</td>
<td class="best">33.0</td>
<td class="best">23.9</td>
</tr>
<tr>
<td style="text-align: left; padding-left: 2em;"><i>Prismatic</i></td>
<td>136</td>
<td>--</td>
<td>--</td>
<td>1.89</td>
<td>8.8</td>
<td>8.1</td>
<td>8.1</td>
<td>8.1</td>
<td>--</td>
<td>--</td>
<td>1.89</td>
<td>8.8</td>
<td>8.1</td>
<td>8.1</td>
<td>8.1</td>
</tr>
<tr>
<td style="text-align: left; padding-left: 2em;"><i>Revolute (Vert.)</i></td>
<td>386</td>
<td>--</td>
<td>0.202</td>
<td>1.10</td>
<td>49.5</td>
<td>46.9</td>
<td>46.9</td>
<td>35.0</td>
<td>--</td>
<td>0.180</td>
<td>1.08</td>
<td>45.6</td>
<td>43.3</td>
<td>43.3</td>
<td>33.4</td>
</tr>
<tr>
<td style="text-align: left; padding-left: 2em;"><i>Revolute (Hor.)</i></td>
<td>18</td>
<td>--</td>
<td>0.316</td>
<td>2.39</td>
<td>50.0</td>
<td>11.1</td>
<td>11.1</td>
<td>5.6</td>
<td>--</td>
<td>--</td>
<td>--</td>
<td>27.8</td>
<td>0.0</td>
<td>0.0</td>
<td>0.0</td>
</tr>
</tbody>
</table>
</div>
</div>
</div>
</div>
</section>
<section class="section" id="Acknowledgements">
<div class="container is-max-desktop content">
<h2 class="title">Acknowledgements</h2>
<p>
This work was supported by the Swiss National Science Foundation Advanced Grant 216260: "Beyond Frozen Worlds: Capturing Functional 3D Digital Twins from the Real World". Alexandros Delitzas is supported by the Max Planck ETH Center for Learning Systems (CLS).
</p>
</div>
</section>
<section class="section" id="BibTeX">
<div class="container is-max-desktop content">
<h2 class="title">BibTeX</h2>
<pre><code>@ARTICLE{11434845,
author={Huang, Zhao and Sun, Boyang and Delitzas, Alexandros and Chen, Jiaqi and Pollefeys, Marc},
journal={IEEE Robotics and Automation Letters},
title={REACT3D: Recovering Articulations for Interactive Physical 3D Scenes},
year={2026},
volume={11},
number={5},
pages={5954-5961},
keywords={Three-dimensional displays;Joints;Geometry;Image reconstruction;Estimation;Solid modeling;Point cloud compression;Foundation models;Biological system modeling;Accuracy;Semantic scene understanding;object detection;segmentation and categorization;RGB-D perception},
doi={10.1109/LRA.2026.3674028}
}</code></pre>
</div>
</section>
<footer class="footer">
<div class="container">
<div class="columns is-centered">
<!-- <div class="column is-8"> -->