Knowing a Good HOG Filter When You See It: Efficient Selection of Filters for Detection thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Knowing a Good HOG Filter When You See It: Efficient Selection of Filters for Detection

Published on Oct 29, 20143605 Views

Collections of filters based on histograms of oriented gradients (HOG) are common for several detection methods, notably, poselets and exemplar SVMs. The main bottleneck in training such systems is th

Related categories

Chapter list

Knowing a Good HOG Filter When You See It:Efficient Selection of Filters for Detection00:00
Visual Category as Collection of filters00:13
Candidate Generation00:40
Candidate Selection01:05
Candidate Selection Cont…02:06
What we Propose02:42
Category Independent Model03:06
Poselets04:04
Poselet Architecture04:20
ESVM05:06
Good / Bad Filters06:00
Features for filter Ranking06:37
Learning to Rank Filters07:35
Greedy approximation for Diversity08:17
LDA Acceleration09:07
Experiments with Poselets10:17
Performance of Ranker10:52
Detection Results11:31
Experiments with exemplar SVMs14:14
Conclusion14:36