Efficient Optimization For Low-Rank Integrated Bilinear Classifiers 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

Efficient Optimization For Low-Rank Integrated Bilinear Classifiers

Published on Nov 12, 20124581 Views

In pattern classification, it is needed to efficiently treat two-way data (feature matrices) while preserving the two-way structure such as spatio-temporal relationships, etc. The classifier for the f

Related categories

Chapter list

Efficient Optimization For Low-Rank Integrated Bilinear Classifiers00:00
Feature Representation in Visual Classification Problems (1)00:21
Feature Representation in Visual Classification Problems (2)00:36
Feature Representation in Visual Classification Problems (3)00:54
Feature Representation in Visual Classification Problems (4)01:05
Feature Representation in Visual Classification Problems (5)01:12
Feature Representation in Visual Classification Problems (6)01:20
Feature Representation in Visual Classification Problems (7)01:27
Feature Representation in Visual Classification Problems (8)01:30
Feature Representation in Visual Classification Problems (9)01:36
Classifier for Feature Matrix01:46
Challenges02:43
Convex Optimization for Bilinear Classifier03:05
Margin of Bilinear Classifier03:09
Related Works04:07
Efficient Convex Optimization (1)04:41
Efficient Convex Optimization (2)05:54
Experimental Results (1)06:16
Experimental Results (2)06:58
Experimental Results (3)07:38
Kernelization - heterogeneous multiple kernel learning08:13
Kernelized Bilinear Model08:15
Heterogeneous Kernel09:15
Connection to Canonical Correlation Analysis (1)10:52
Connection to Canonical Correlation Analysis (2)11:24
Connection to Canonical Correlation Analysis (3)11:50
eccv2012_kobayashi_optimization_01_Page_2711:56
eccv2012_kobayashi_optimization_01_Page_2812:06
eccv2012_kobayashi_optimization_01_Page_2912:13
Connection to Canonical Correlation Analysis (4)12:41
Recap: Heterogeneous MKL (1)13:03
Recap: Heterogeneous MKL (2)13:12
Experimental Results (4)13:42
Experimental Results (5)14:24
Conclusion14:47
Thank you!!14:59