Mask-guided contrastive attention and two-stream metric co-learning for person Re-identification

doi:10.1016/j.neucom.2021.09.038

CORC > 自动化研究所 > 中国科学院自动化研究所 > 智能感知与计算研究中心

	Mask-guided contrastive attention and two-stream metric co-learning for person Re-identification
	Song, Chunfeng 1,2; Shan, Caifeng 3,4; Huang, Yan 1,2; Wang, Liang 1,2,4
刊名	NEUROCOMPUTING
	2021-11-20
卷号	465 页码:561-573
关键词	Person ReID Contrastive attention model Two-stream metric learning
ISSN号	0925-2312
DOI	10.1016/j.neucom.2021.09.038
通讯作者	Shan, Caifeng(caifeng.shan@gmail.com)
英文摘要	Person Re-identification (ReID) is an important yet challenging task in computer vision. Due to diverse background clutters, variations of viewpoints and body poses, it is far from being solved. How to extract discriminative and robust features invariant to background clutters is one of the core problems. In this paper, we first introduce a set of binary segmentation masks to construct synthetic RGB-Mask pairs as inputs, and then design a mask-guided contrastive attention model (MGCAM) to learn features separately from the body and background regions. Moreover, we propose a novel region-level triplet loss to guide the features learning, i.e., pulling the features from the full image and body region close, whereas pushing the features from backgrounds away. To learn the similarities from multiple features of the proposed MGCAM, we further introduce the instance-level two-stream metric co-learning (TSMCL) to help learn pair-wise relations between features from not only different regions but also different instances. TSMCL could help learn more compact features across full and body streams, enhancing the performance of MGCAM. We evaluate the proposed method on four public datasets, including MARS, Market-1501, CUHK03, and DukeMTMC-reID. Extensive experiments show that the proposed method is effective and achieves satisfying results. (c) 2021 Elsevier B.V. All rights reserved.
资助项目	National Natural Science Foundation of China[62006231] ; National Natural Science Foundation of China[61525306] ; National Natural Science Foundation of China[61633021] ; National Natural Science Foundation of China[61721004] ; National Natural Science Foundation of China[61420106015] ; National Key Research and Development Program of China[2016YFB1001000] ; Shandong Provincial Key Research and Development Program (Major Scientific and Technological Innovation Project)[2019JZZY010-119] ; Capital Science and Technology Leading Talent Training Project[Z181100006318030] ; NVIDIA ; NVIDIA DGX-1 AI Supercomputer
WOS关键词	NETWORK
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000702819100009
资助机构	National Natural Science Foundation of China ; National Key Research and Development Program of China ; Shandong Provincial Key Research and Development Program (Major Scientific and Technological Innovation Project) ; Capital Science and Technology Leading Talent Training Project ; NVIDIA ; NVIDIA DGX-1 AI Supercomputer
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/45735]
专题	自动化研究所_智能感知与计算研究中心
通讯作者	Shan, Caifeng
作者单位	1.Chinese Acad Sci CASIA, Inst Automat, Ctr Res Intelligent Percept & Comp CRIPAC, Natl Lab Pattern Recognit NLPR, Beijing, Peoples R China 2.Univ Chinese Acad Sci UCAS, Beijing, Peoples R China 3.Shandong Univ Sci & Technol, Qingdao, Peoples R China 4.CAS CAS AIR, Artificial Intelligence Res, Qingdao, Peoples R China
推荐引用方式 GB/T 7714	Song, Chunfeng,Shan, Caifeng,Huang, Yan,et al. Mask-guided contrastive attention and two-stream metric co-learning for person Re-identification[J]. NEUROCOMPUTING,2021,465:561-573.
APA	Song, Chunfeng,Shan, Caifeng,Huang, Yan,&Wang, Liang.(2021).Mask-guided contrastive attention and two-stream metric co-learning for person Re-identification.NEUROCOMPUTING,465,561-573.
MLA	Song, Chunfeng,et al."Mask-guided contrastive attention and two-stream metric co-learning for person Re-identification".NEUROCOMPUTING 465(2021):561-573.