Dynamic Re-Weighting and Cross-Camera Learning for Unsupervised Person Re-Identification

doi:10.3390/math10101654

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 生物识别与安全技术

	Dynamic Re-Weighting and Cross-Camera Learning for Unsupervised Person Re-Identification
	Yin, Qingze 1; Wang, Guan'an 2; Wu, Jinlin2 ; Luo, Haonan 1; Tang, Zhenmin 1
发表期刊	MATHEMATICS
	2022-05-01
卷号	10 期号:10 页码:17
通讯作者	Tang, Zhenmin(tzm.cs@njust.edu.cn)
摘要	Person Re-Identification (ReID) has witnessed tremendous improvements with the help of deep convolutional neural networks (CNN). Nevertheless, because different fields have their characteristics, most existing methods encounter the problem of poor generalization ability to invisible people. To address this problem, based on the relationship between the temporal and camera position, we propose a robust and effective training strategy named temporal smoothing dynamic re-weighting and cross-camera learning (TSDRC). It uses robust and effective algorithms to transfer valuable knowledge of existing labeled source domains to unlabeled target domains. In the target domain training stage, TSDRC iteratively clusters the samples into several centers and dynamically re-weights unlabeled samples from each center with a temporal smoothing score. Then, cross-camera triplet loss is proposed to fine-tune the source domain model. Particularly, to improve the discernibility of CNN models in the source domain, generally shared person attributes and margin-based softmax loss are adapted to train the source model. In terms of the unlabeled target domain, the samples are clustered into several centers iteratively and the unlabeled samples are dynamically re-weighted from each center. Then, cross-camera triplet loss is proposed to fine-tune the source domain model. Comprehensive experiments on the Market-1501 and DukeMTMC-reID datasets demonstrate that the proposed method vastly improves the performance of unsupervised domain adaptability.
关键词	clustering dynamic re-weighting person attributes cross-camera triplet loss
DOI	10.3390/math10101654
关键词[WOS]	ATTRIBUTE ; FEATURES
收录类别	SCI
语种	英语
WOS研究方向	Mathematics
WOS类目	Mathematics
WOS记录号	WOS:000803416800001
出版者	MDPI
引用统计	被引频次：2[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/49537
专题	多模态人工智能系统全国重点实验室_生物识别与安全技术
通讯作者	Tang, Zhenmin
作者单位	1.Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, 200 Xiaolingwei St, Nanjing 210094, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Yin, Qingze,Wang, Guan'an,Wu, Jinlin,et al. Dynamic Re-Weighting and Cross-Camera Learning for Unsupervised Person Re-Identification[J]. MATHEMATICS,2022,10(10):17.
APA	Yin, Qingze,Wang, Guan'an,Wu, Jinlin,Luo, Haonan,&Tang, Zhenmin.(2022).Dynamic Re-Weighting and Cross-Camera Learning for Unsupervised Person Re-Identification.MATHEMATICS,10(10),17.
MLA	Yin, Qingze,et al."Dynamic Re-Weighting and Cross-Camera Learning for Unsupervised Person Re-Identification".MATHEMATICS 10.10(2022):17.