en
0.25
0.5
0.75
1.25
1.5
1.75
2
Constrained Semi-Supervised Learning using Attributes and Comparative Attributes
Published on Nov 12, 20126558 Views
We consider the problem of semi-supervised bootstrap learning for scene categorization. Existing semi-supervised approaches are typically unreliable and face semantic drift because the learning task
Related categories
Chapter list
Constrained Semi-Supervised Learning Using Attributes and Comparative Attributes00:00
Big-data00:03
Unsupervised00:36
Semi-Supervised00:53
Semi-Supervised Bootstrapping (1)01:03
Semi-Supervised Bootstrapping (2)01:45
Semi-Supervised Bootstrapping (3)01:50
Semi-Supervised Bootstrapping (4)01:54
Semi-Supervised Graph-baesd Methods02:19
Our Approach (1)02:48
Our Approach (2)03:05
Binary Attributes (BA) (1)03:20
Binary Attributes (BA) (2)03:28
Binary Attributes (BA) (3)03:33
Auditorium (1)03:50
Auditorium (2)03:54
Auditorium (3)04:07
Sharing via Dissimilarity04:22
Amphitheatre, Auditorium (1)04:39
Amphitheatre, Auditorium (2)04:46
Amphitheatre, Auditorium (3)04:54
Dissimilarity Comparative Attributes (1)05:09
Dissimilarity Comparative Attributes (2)05:18
Dissimilarity Comparative Attributes (3)05:29
Comparative Attributes05:38
Selected Candidates (1)06:08
Selected Candidates (2)06:11
Promoted Instances06:15
Introspection07:11
Amphitheatre07:55
Experimental Evaluation08:27
Control Experiments (1)08:29
Control Experiments (2)08:57
Control Experiments (3)09:03
Control Experiments (4)09:10
Control Experiments (5)09:14
Control Experiments (6)09:21
Control Experiments (7)09:32
Banquet Hall09:40
Control Experiments10:07
Co-training (Small Scale)10:14
Bedroom10:22
Scene Classification10:38
Co-Training (Large Scale)10:45
Conclusion11:03
Thank you!11:25