en
0.25
0.5
0.75
1.25
1.5
1.75
2
Semantic Segmentation with Second-Order Pooling
Published on 2012-11-125948 Views
Feature extraction, coding and pooling, are important components on many contemporary object recognition paradigms. In this paper we explore novel pooling techniques that encode the second-order stati
Related categories
Presentation
Semantic Segmentation with Second-Order Pooling00:00
Semantic Segmentation (1)00:05
Semantic Segmentation (2)00:39
Semantic Segmentation (3)00:55
Semantic Segmentation (4)01:24
Semantic Segmentation (5)01:31
Semantic Segmentation (6)01:40
Semantic Segmentation (7)01:42
Semantic Segmentation (8)01:43
Semantic Segmentation (9)01:44
Semantic Segmentation (10)01:52
Describing Free-form Regions01:53
Aggregation-based Descriptors (1)02:26
Aggregation-based Descriptors (2)02:57
Aggregation-based Descriptors (3)03:02
Aggregation-based Descriptors (4)03:18
Aggregation-based Descriptors (5)03:28
Aggregation-based Descriptors (6)03:39
Aggregation-based Descriptors (7)03:58
Second-Order Pooling (1)04:28
Second-Order Pooling (2)04:32
Second-Order Pooling (3)04:41
Second-Order Pooling (4)04:59
Second-Order Pooling (5)05:12
Second-Order Pooling (6)05:18
Second-Order Pooling (7)05:23
Second-Order Pooling (8)05:41
Second-Order Pooling (9)05:45
Embedding SPD Manifold in Euclidean Space (1)05:51
Embedding SPD Manifold in Euclidean Space (2)06:05
Sequence of Operations (1)06:30
Sequence of Operations (2)07:15
Additionally we use better local descriptors with pooling methods07:17
Local Feature Enrichment (1/2) Relative Position07:34
Local Feature Enrichment (2/2) Pixel Color08:01
Region Classification VOC 2011 (1)08:17
Region Classification VOC 2011 (2)08:59
Region Classification VOC 2011 (3)09:08
Region Classification VOC 2011 (4)09:31
Region Classification VOC 2011 (5)09:47
Semantic Segmentation in the Wild Pascal VOC 2011 (1)09:56
Semantic Segmentation in the Wild Pascal VOC 2011 (2)11:08
Semantic Segmentation in the Wild Pascal VOC 2011 (3)11:56
Semantic Segmentation in the Wild Pascal VOC 2011 (4)12:04
Caltech 101 (1)12:27
Caltech 101 (2)12:38
Caltech 101 (3)12:44
Conclusions13:03
Thank you!14:04