CASIA OpenIR
GaitGANv2: Invariant gait feature extraction using generative adversarial networks
Yu, Shiqi1; Liao, Rijun1; An, Weizhi1; Chen, Haifeng1; Garcia, Edel B.2; Huang, Yongzhen3,4; Poh, Norman5
Source PublicationPATTERN RECOGNITION
ISSN0031-3203
2019-03-01
Volume87Pages:179-189
Corresponding AuthorYu, Shiqi(shiqi.yu@szu.edu.cn)
AbstractThe performance of gait recognition can be adversely affected by many sources of variation such as view angle, clothing, presence of and type of bag, posture, and occlusion, among others. To extract invariant gait features, we proposed a method called GaitGANv2 which is based on generative adversarial networks (GAN). In the proposed method, a GAN model is taken as a regressor to generate a canonical side view of a walking gait in normal clothing without carrying any bag. A unique advantage of this approach is that, unlike other methods, GaitGANv2 does not need to determine the view angle before generating invariant gait images. Indeed, only one model is needed to account for all possible sources of variation such as with or without carrying accessories and varying degrees of view angle. The most important computational challenge, however, is to address how to retain useful identity information when generating the invariant gait images. To this end, our approach differs from the traditional GAN in that GaitGANv2 contains two discriminators instead of one. They are respectively called fake/real discriminator and identification discriminator. While the first discriminator ensures that the generated gait images are realistic, the second one maintains the human identity information. The proposed GaitGANv2 represents an improvement over GaitGANvl in that the former adopts a multi-loss strategy to optimize the network to increase the inter-class distance and to reduce the intra-class distance, at the same time. Experimental results show that GaitGANv2 can achieve state-of-the-art performance. (C) 2018 Elsevier Ltd. All rights reserved.
KeywordGait recognition Generative adversarial networks Invariant feature
DOI10.1016/j.patcog.2018.10.019
WOS KeywordRECOGNITION ; PROJECTION
Indexed BySCI
Language英语
Funding ProjectScience Foundation of Shenzhen[JCYJ20150324141711699] ; Science Foundation of Shenzhen[20170504160426188]
Funding OrganizationScience Foundation of Shenzhen
WOS Research AreaComputer Science ; Engineering
WOS SubjectComputer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS IDWOS:000453338200015
PublisherELSEVIER SCI LTD
Citation statistics
Cited Times:6[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/25662
Collection中国科学院自动化研究所
Corresponding AuthorYu, Shiqi
Affiliation1.Shenzhen Univ, Coll Comp Sci & Software Engn, Shenzhen, Peoples R China
2.Adv Technol Applicat Ctr CENATAV, Havana, Cuba
3.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China
4.Watrix Technol Ltd Co Ltd, Suzhou, Peoples R China
5.Trust Stamp, Atlanta, GA USA
Recommended Citation
GB/T 7714
Yu, Shiqi,Liao, Rijun,An, Weizhi,et al. GaitGANv2: Invariant gait feature extraction using generative adversarial networks[J]. PATTERN RECOGNITION,2019,87:179-189.
APA Yu, Shiqi.,Liao, Rijun.,An, Weizhi.,Chen, Haifeng.,Garcia, Edel B..,...&Poh, Norman.(2019).GaitGANv2: Invariant gait feature extraction using generative adversarial networks.PATTERN RECOGNITION,87,179-189.
MLA Yu, Shiqi,et al."GaitGANv2: Invariant gait feature extraction using generative adversarial networks".PATTERN RECOGNITION 87(2019):179-189.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Yu, Shiqi]'s Articles
[Liao, Rijun]'s Articles
[An, Weizhi]'s Articles
Baidu academic
Similar articles in Baidu academic
[Yu, Shiqi]'s Articles
[Liao, Rijun]'s Articles
[An, Weizhi]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Yu, Shiqi]'s Articles
[Liao, Rijun]'s Articles
[An, Weizhi]'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.