Isosceles Constraints for Person Re-Identification

doi:10.1109/TIP.2020.3020648

	Isosceles Constraints for Person Re-Identification
	Xu, Furong 1; Ma, Bingpeng 1,2; Chang, Hong 2,3; Shan, Shiguang 2,3,4
	2020
发表期刊	IEEE TRANSACTIONS ON IMAGE PROCESSING
ISSN	1057-7149
卷号	29 页码:8930-8943
摘要	In the existing works of person re-identification (ReID), batch hard triplet loss has achieved great success. However, it only cares about the hardest samples within the batch. For any probe, there are massive mismatched samples (crucial samples) outside the batch which are closer than the matched samples. To reduce the disruptive influence of crucial samples, we propose a novel isosceles contraint for triplet. Theoretically, we show that if a matched pair has equal distance to any one of mismatched sample, the matched pair should be infinitely close. Motivated by this, the isosceles constraint is designed for the two mismatched pairs of each triplet, to restrict some matched pairs with equal distance to different mismatched samples. Meanwhile, to ensure that the distance of mismatched pairs are larger than the matched pairs, margin constraints are necessary. Minimizing the isosceles and margin constraints with respect to the feature extraction network makes the matched pairs closer and the mismatched pairs farther away than the matched ones. By this way, crucial samples are effectively reduced and the performance on ReID is improved greatly. Likewise, our isosceles contraint can be applied to quadruplet as well. Comprehensive experimental evaluations on Market-1501, DukeMTMC-reID and CUHK03 datasets demonstrate the advantages of our isosceles constraint over the related state-of-the-art approaches.
关键词	Feature extraction Training Measurement Robustness Machine learning Probes Task analysis Person re-identification isosceles constraint triplet quadruplet
DOI	10.1109/TIP.2020.3020648
收录类别	SCI
语种	英语
资助项目	Natural Science Foundation of China (NSFC)[61732004] ; Natural Science Foundation of China (NSFC)[61876171] ; Natural Science Foundation of China (NSFC)[61976203] ; Fundamental Research Funds for the Central Universities
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000572623700001
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：12[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/15589
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Ma, Bingpeng
作者单位	1.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 100049, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 3.Univ Chinese Acad Sci, Sch Comp Sci & Technol, Beijing 100049, Peoples R China 4.CAS Ctr Excellence Brain Sci & Intelligence Techn, Shanghai 200031, Peoples R China
推荐引用方式 GB/T 7714	Xu, Furong,Ma, Bingpeng,Chang, Hong,et al. Isosceles Constraints for Person Re-Identification[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2020,29:8930-8943.
APA	Xu, Furong,Ma, Bingpeng,Chang, Hong,&Shan, Shiguang.(2020).Isosceles Constraints for Person Re-Identification.IEEE TRANSACTIONS ON IMAGE PROCESSING,29,8930-8943.
MLA	Xu, Furong,et al."Isosceles Constraints for Person Re-Identification".IEEE TRANSACTIONS ON IMAGE PROCESSING 29(2020):8930-8943.