CASIA OpenIR  > 模式识别国家重点实验室  > 模式分析与学习
SDE: A novel selective, discriminative and equalizing feature representation for visual recognition
Guo-Sen Xie1; Xu-Yao Zhang1; Shuicheng Yan2; Cheng-Lin Liu1
Source PublicationInternational Journal of Computer Vision (IJCV)
2017
Volume124Issue:2Pages:145~168
AbstractBag of Words (Bo W) model and Convolutional Neural Network (CNN) are two milestones in visual recognition. Both Bo W and CNN require a feature pooling operation for constructing the frameworks. Particularly, the max-pooling has been validated as an efficient and effective pooling method compared with other methods such as average pooling and stochastic pooling. In this paper, we first evaluate different pooling methods, and then propose a new feature pooling method termed as selective, discriminative and equalizing pooling (SDE). The SDE representation is a feature learning mechanism by jointly optimizing the pooled representations with the target of learning more selective, discriminative and equalizing features. We use bilevel optimization to solve the joint optimization problem. Experiments on seven benchmark datasets (including both singlelabel and multi-label ones) well validate the effectiveness of our framework. Particularly, we achieve the state-of-the-art fused results (m AP) of 93.21 and 93.97% on the PASCAL VOC2007 and VOC2012 datasets, respectively.
KeywordConvolutional Neural Network · Feature Learning · Pooling · Bag Of Words · Bilevel Optimization
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/20355
Collection模式识别国家重点实验室_模式分析与学习
Corresponding AuthorCheng-Lin Liu
Affiliation1.Institute of automation, Chinese Academy of Sciences
2.Department of Electrical and Computer Engineering, National University of Singapore
Recommended Citation
GB/T 7714
Guo-Sen Xie,Xu-Yao Zhang,Shuicheng Yan,et al. SDE: A novel selective, discriminative and equalizing feature representation for visual recognition[J]. International Journal of Computer Vision (IJCV),2017,124(2):145~168.
APA Guo-Sen Xie,Xu-Yao Zhang,Shuicheng Yan,&Cheng-Lin Liu.(2017).SDE: A novel selective, discriminative and equalizing feature representation for visual recognition.International Journal of Computer Vision (IJCV),124(2),145~168.
MLA Guo-Sen Xie,et al."SDE: A novel selective, discriminative and equalizing feature representation for visual recognition".International Journal of Computer Vision (IJCV) 124.2(2017):145~168.
Files in This Item: Download All
File Name/Size DocType Version Access License
SDE A novel selectiv(5169KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Guo-Sen Xie]'s Articles
[Xu-Yao Zhang]'s Articles
[Shuicheng Yan]'s Articles
Baidu academic
Similar articles in Baidu academic
[Guo-Sen Xie]'s Articles
[Xu-Yao Zhang]'s Articles
[Shuicheng Yan]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Guo-Sen Xie]'s Articles
[Xu-Yao Zhang]'s Articles
[Shuicheng Yan]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: SDE A novel selective, discriminative and equalizing feature representation for visual recognition.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.