Indoor Segmentation and Support Inference from RGBD Images thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Indoor Segmentation and Support Inference from RGBD Images

Published on Nov 12, 20126125 Views

We present an approach to interpret the major surfaces, objects, and support relations of an indoor scene from an RGBD image. Most existing work ignores physical interactions or is applied only to tid

Related categories

Chapter list

Indoor Segmentation and Support Inference from RGBD Images00:00
Goal: Infer Support for Every Region (1)00:04
Goal: Infer Support for Every Region (2)00:15
Why infer physical support? 00:24
Why infer physical support: Recognition (1)00:43
Why infer physical support: Recognition (2)00:58
Working with RGB+Depth01:00
NYU Depth Dataset Version 2.001:18
High Quality Semantic Labels01:45
High Quality Support Labels02:02
RGBD Image -> Segmentation -> Support Inference02:14
Scene Parsing (1)02:23
Scene Parsing (2)02:29
Scene Parsing (302:41
Hierarchical Segmentation (1)02:47
Hierarchical Segmentation (2)03:02
Hierarchical Segmentation (3)03:04
Hierarchical Segmentation (4)03:07
Scene Parsing (4)03:13
Scene Parsing (5)03:15
RGBD Image -> Segmentation -> Support Inference (1)03:21
Modeling Choice #103:23
Modeling Choice #203:43
Modeling Choice #2a03:46
Modeling Choice #2b03:53
Modeling Choice #3a03:59
Modeling Choice #3b04:02
Modeling Choice #3c04:05
eccv2012_silberman_images_01_Page_3004:08
Modeling Support: Structure Classes04:48
Modeling Support (1)05:20
Modeling Support (2)05:31
Modeling Support (3)05:41
Modeling Support (4)06:04
Modeling Support (5)06:19
Modeling Support (6)06:41
Modeling Support (7)06:48
Modeling Support (8)07:01
Integer Program Formulation07:20
Experiments07:47
Evaluating Support (1)07:48
Baseline #1: Image Plane Rules08:17
Baselines #2: Structure Class Rules08:30
Baselines #3: Support Classifier08:44
Evaluating Support (2)08:53
Results (1)09:49
Results (2)10:05
Results (3)10:20
Results (4)10:21
Results (5)10:23
Results (6)10:26
Results (7)10:29
Results (8)10:32
Results (9)11:01
Evaluating Support (3)11:36
Results (10)12:09
Results (11)12:49
Conclusion13:09