en-es
en
0.25
0.5
0.75
1.25
1.5
1.75
2
Unsupervised Object Discovery and Segmentation in Videos
Published on Apr 03, 20142955 Views
Unsupervised object discovery is the task of finding recurring objects over an unsorted set of images without any human supervision, which becomes more and more important as the amount of visual dat
Related categories
Chapter list
Unsupervised Object Discovery and Segmentation in Videos00:00
What is Unsupervised Object Discovery? - 100:12
What is Unsupervised Object Discovery? - 200:19
What is Unsupervised Object Discovery? - 300:27
What is Unsupervised Object Discovery? - 400:28
Typical Approach00:30
Use Videos instead of Still Images01:17
UOD in Videos01:55
Outline02:16
Building Blocks - 102:33
Building Blocks - 202:40
Building Blocks - 302:46
Building Blocks - 403:00
Building Blocks - 503:17
Building Blocks - 603:20
Motion Segmentation - 103:45
Motion Segmentation - 204:13
Motion Segmentation - 304:19
Object Proposals from Motion04:30
Object Proposal Clustering05:37
Clustering Result - 106:41
Clustering Result - 207:08
Training Object Models07:15
Applying Object Models08:26
Recap - 109:01
Recap - 209:23
Recap - 309:39
CRF-based Semantic Segmentation - 109:45
CRF-based Semantic Segmentation - 210:49
CRF-based Semantic Segmentation - 312:07
Experiments12:26
Object Discovery in Videos13:03
Quantitative Results - 113:48
Quantitative Results - 214:46
Quantitative Results - 314:59
Result Videos15:14
Recognition in Still Images - 115:24
Recognition in Still Images - 215:37
Recognition in Still Images - 315:45
Results on TUD-pedestrian16:25
Results on ETHZ-cars17:07
Conclusion17:35
Thank you!18:15