DMCA
Semantic segmentation with second-order pooling
Cached
Download Links
| Venue: | In ECCV |
| Citations: | 91 - 12 self |
BibTeX
@INPROCEEDINGS{Carreira_semanticsegmentation,
author = {João Carreira and Jorge Batista and Cristian Sminchisescu},
title = {Semantic segmentation with second-order pooling},
booktitle = {In ECCV},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract. 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 statistics of local descriptors inside a region. To achieve this effect, we introduce multiplicative second-order analogues of average and maxpooling that together with appropriate non-linearities lead to state-ofthe-art performance on free-form region recognition, without any type of feature coding. Instead of coding, we found that enriching local descriptors with additional image information leads to large performance gains, especially in conjunction with the proposed pooling methodology. We show that second-order pooling over free-form regions produces results superior to those of the winning systems in the Pascal VOC 2011 semantic segmentation challenge, with models that are 20,000 times faster.
Keyphrases
second-order pooling semantic segmentation local descriptor free-form region second-order statistic multiplicative second-order analogue semantic segmentation challenge feature extraction important component pascal voc appropriate non-linearities lead state-ofthe-art performance large performance gain many contemporary object recognition paradigm pooling methodology additional image information free-form region recognition
