Sign inSign up
Sliding Shapes for 3D Object Detection in Depth Images thumbnail
slide-image
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Sliding Shapes for 3D Object Detection in Depth Images

Published on Oct 29, 201415170 Views

Shuran Song

The depth information of RGB-D sensors has greatly simplified some common challenges in computer vision and enabled breakthroughs for several tasks. In this paper, we propose to use depth maps for obj

ECCV 2014 - Zurich

Related categories

Computer Vision

Chapter list

Sliding Shapes for 3D Object Detection in Depth Images00:00
Object Detection - 100:12
Object Detection - 200:25
State-of-the-Art: Deep Learning00:28
NYU Algorithm on NYU Dataset - 100:43
NYU Algorithm on NYU Dataset - 201:01
NYU Algorithm on NYU Dataset - 301:20
NYU Algorithm on NYU Dataset - 401:32
Object Detection is Hard01:42
Maybe Depth Sensors can Help?02:36
Depth for Other Vision Tasks02:53
Sliding Shapes03:08
Depth-based Object Detection - 103:22
Depth-based Object Detection - 203:25
Algorithm03:34
Training: CADmodels - 103:41
Training: CADmodels - 203:51
Training: Rendering Depth03:52
Training: 3D Exemplar SVM04:04
Recipe of Detector Features04:19
3D Features04:45
Testing: 3D Sliding Window - 105:02
Testing: 3D Sliding Window - 205:28
Testing: 3D Sliding Window - 305:29
Testing: 3D Sliding Window - 405:30
Testing: 3D Sliding Window - 505:32
Testing: 3D Sliding Window - 605:33
Testing: 3D Sliding Window - 705:37
Testing: 3D Sliding Window - 805:44
Results - 105:50
Results - 206:12
Results - 306:20
Evaluation06:30
Comparison with DPM - 106:48
Comparison with DPM - 207:02
Comparison with DPM - 307:12
Comparison with DPM - 407:25
Comparison with DPM - 507:35
False Positives - 107:55
False Positives - 208:12
Analysis08:20
Object Detection is Hard - 108:27
Object Detection is Hard - 208:38
Problem: Intra-class Variations08:39
Solution: Data-driven Exemplar08:50
Problem: Viewpoints09:06
Solution: Numerate All Views - 109:15
Solution: Numerate All Views - 209:18
Solution: Numerate All Views - 309:26
Problem: Illumination09:32
Solution: 3D Depth09:53
Problem: Clutter10:10
Solution: Occupation Mask - 110:27
Solution: Occupation Mask - 210:37
Solution: Occupation Mask - 310:53
Problem: Occlusion11:05
Solution: 3D Window11:14
Data11:44
Quantity & Quality11:49
CG vs. Kinect - 112:33
CG vs. Kinect - 212:57
CG Rendering as Negatives13:33
Conclusion14:04
Usage of Kinect14:09
Sliding Shapes - 114:12
Sliding Shapes - 217:28
Time complexity and time17:28