SIA OpenIR  > 机器人学研究室
Fine-Grained Spatial Alignment Model for Person Re-Identification with Focal Triplet Loss
Zhou, Qinqin1; Zhong BN(钟必能)1; Lan, Xiangyuan2; Sun G(孙干)3; Zhang, Yulun4; Zhang, Baochang5; Ji RR(纪荣嵘)6
Department机器人学研究室
Source PublicationIEEE Transactions on Image Processing
ISSN1057-7149
2020
Volume29Pages:7578-7589
Indexed BySCI ; EI
EI Accession number20203008971440
WOS IDWOS:000553851400025
Contribution Rank3
Funding OrganizationNational Natural Science Foundation of China under Grant U1705262, Grant 61972167, and Grant 61802135 ; National Key Research and Development Program under Grant 2017YFC0113000 and Grant 2016YFB1001503 ; Fundamental Research Funds for the Central Universities under Grant 30918014108 ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR) under Grant 202000012
KeywordPerson re-identification spatial alignment focal triplet loss
Abstract

Recent advances of person re-identification have well advocated the usage of human body cues to boost performance. However, most existing methods still retain on exploiting a relatively coarse-grained local information. Such information may include redundant backgrounds that are sensitive to the apparently similar persons when facing challenging scenarios like complex poses, inaccurate detection, occlusion and misalignment. In this paper we propose a novel Fine-Grained Spatial Alignment Model (FGSAM) to mine fine-grained local information to handle the aforementioned challenge effectively. In particular, we first design a pose resolve net with channel parse blocks (CPB) to extract pose information in pixel-level. This network allows the proposed model to be robust to complex pose variations while suppressing the redundant backgrounds caused by inaccurate detection and occlusion. Given the extracted pose information, a locally reinforced alignment mode is further proposed to address the misalignment problem between different local parts by considering different local parts along with attribute information in a fine-grained way. Finally, a focal triplet loss is designed to effectively train the entire model, which imposes a constraint on the intra-class and an adaptively weight adjustment mechanism to handle the hard sample problem. Extensive evaluations and analysis on Market1501, DukeMTMC-reid and PETA datasets demonstrate the effectiveness of FGSAM in coping with the problems of misalignment, occlusion and complex poses.

Language英语
WOS SubjectComputer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS KeywordTRACKING
WOS Research AreaComputer Science ; Engineering
Funding ProjectNational Natural Science Foundation of China[U1705262] ; National Natural Science Foundation of China[61972167] ; National Natural Science Foundation of China[61802135] ; National Key Research and Development Program[2017YFC0113000] ; National Key Research and Development Program[2016YFB1001503] ; Fundamental Research Funds for the Central Universities[30918014108] ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR)[202000012]
Citation statistics
Cited Times:24[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/27367
Collection机器人学研究室
Corresponding AuthorZhong BN(钟必能); Ji RR(纪荣嵘)
Affiliation1.Department of Computer Science and Technology, Huaqiao University, Xiamen, China
2.Department of Computer Science, Hong Kong Baptist University, Hong Kong
3.State Key Laboratory of Robotics, Shenyang Institute of Automation, Chinese Academy of Sciences, Shenyang, China
4.Department of ECE, Northeastern University, Boston
5.MA, United States
6.School of Automation Science and Electrical Engineering, Beihang University, Beijing, China
7.Department of Artificial Intelligence, School of Informatics, Media Analytics and Computing Laboratory, Xiamen University, Xiamen, China
Recommended Citation
GB/T 7714
Zhou, Qinqin,Zhong BN,Lan, Xiangyuan,et al. Fine-Grained Spatial Alignment Model for Person Re-Identification with Focal Triplet Loss[J]. IEEE Transactions on Image Processing,2020,29:7578-7589.
APA Zhou, Qinqin.,Zhong BN.,Lan, Xiangyuan.,Sun G.,Zhang, Yulun.,...&Ji RR.(2020).Fine-Grained Spatial Alignment Model for Person Re-Identification with Focal Triplet Loss.IEEE Transactions on Image Processing,29,7578-7589.
MLA Zhou, Qinqin,et al."Fine-Grained Spatial Alignment Model for Person Re-Identification with Focal Triplet Loss".IEEE Transactions on Image Processing 29(2020):7578-7589.
Files in This Item:
File Name/Size DocType Version Access License
Fine-Grained Spatial(1736KB)期刊论文出版稿开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Zhou, Qinqin]'s Articles
[Zhong BN(钟必能)]'s Articles
[Lan, Xiangyuan]'s Articles
Baidu academic
Similar articles in Baidu academic
[Zhou, Qinqin]'s Articles
[Zhong BN(钟必能)]'s Articles
[Lan, Xiangyuan]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Zhou, Qinqin]'s Articles
[Zhong BN(钟必能)]'s Articles
[Lan, Xiangyuan]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Fine-Grained Spatial Alignment Model for Person Re-Identification with Focal Triplet Loss.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

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