UM  > 科技學院  > 電腦及資訊科學系
Person Re-Identification by Dual-Regularized KISS Metric Learning
Dapeng Tao1; Yanan Guo2; Mingli Song3; Yaotang Li2; Zhengtao Yu4; Yuan Yan Tang5,6
2016-06
Source PublicationIEEE Transactions on Image Processing
ISSN1057-7149
Volume25Issue:6Pages:2726 - 2738
Abstract

Person re-identification aims to match the images of pedestrians across different camera views from different locations. This is a challenging intelligent video surveillance problem that remains an active area of research due to the need for performance improvement. Person re-identification involves two main steps: feature representation and metric learning. Although the keep it simple and straightforward (KISS) metric learning method for discriminative distance metric learning has been shown to be effective for the person re-identification, the estimation of the inverse of a covariance matrix is unstable and indeed may not exist when the training set is small, resulting in poor performance. Here, we present dual-regularized KISS (DR-KISS) metric learning. By regularizing the two covariance matrices, DR-KISS improves on KISS by reducing overestimation of large eigenvalues of the two estimated covariance matrices and, in doing so, guarantees that the covariance matrix is irreversible. Furthermore, we provide theoretical analyses for supporting the motivations. Specifically, we first prove why the regularization is necessary. Then, we prove that the proposed method is robust for generalization. We conduct extensive experiments on three challenging person re-identification datasets, VIPeR, GRID, and CUHK 01, and show that DR-KISS achieves new state-of-the-art performance. 

KeywordIntelligent Video Surveillance Metric Learning Person Re-identification Regularization Technique
DOIhttps://doi.org/10.1109/TIP.2016.2553446
URLView the original
Indexed BySCI
Language英语
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS IDWOS:000375472600007
PublisherIEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC, 445 HOES LANE, PISCATAWAY, NJ 08855-4141 USA
The Source to ArticleScopus
Fulltext Access
Citation statistics
Cited Times [WOS]:84   [WOS Record]     [Related Records in WOS]
Document TypeJournal article
CollectionDEPARTMENT OF COMPUTER AND INFORMATION SCIENCE
Corresponding AuthorDapeng Tao; Yanan Guo; Mingli Song; Yaotang Li; Zhengtao Yu; Yuan Yan Tang
Affiliation1.School of Information Science and Engineering, Yunnan University, Kunming 650091, China
2.School of Mathematics and Statistics, Yunnan University, Kunming 650091, China
3.College of Computer Science, Zhejiang University, Hangzhou 310027, China
4.School of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650091, China
5.Faculty of Science and Technology, University of Macau, Macau 999078, China,
6.College of Computer Science, Chongqing University, Chongqing 400000, China
Corresponding Author AffilicationFaculty of Science and Technology
Recommended Citation
GB/T 7714
Dapeng Tao,Yanan Guo,Mingli Song,et al. Person Re-Identification by Dual-Regularized KISS Metric Learning[J]. IEEE Transactions on Image Processing,2016,25(6):2726 - 2738.
APA Dapeng Tao,Yanan Guo,Mingli Song,Yaotang Li,Zhengtao Yu,&Yuan Yan Tang.(2016).Person Re-Identification by Dual-Regularized KISS Metric Learning.IEEE Transactions on Image Processing,25(6),2726 - 2738.
MLA Dapeng Tao,et al."Person Re-Identification by Dual-Regularized KISS Metric Learning".IEEE Transactions on Image Processing 25.6(2016):2726 - 2738.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Dapeng Tao]'s Articles
[Yanan Guo]'s Articles
[Mingli Song]'s Articles
Baidu academic
Similar articles in Baidu academic
[Dapeng Tao]'s Articles
[Yanan Guo]'s Articles
[Mingli Song]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Dapeng Tao]'s Articles
[Yanan Guo]'s Articles
[Mingli Song]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.
 

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