Deep Graph Metric Learning for Weakly Supervised Person Re-Identification

doi:10.1109/TPAMI.2021.3084613

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

	Deep Graph Metric Learning for Weakly Supervised Person Re-Identification
	Meng, Jingke 1,2; Zheng, Wei-Shi 3,5; Lai, Jian-Huang 1; Wang, Liang 4
刊名	IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
	2022-10-01
卷号	44 期号:10 页码:6074-6093
关键词	Training Cameras Labeling Probes Visualization Annotations Loss measurement Person re-identification weakly supervised person re-identification visual surveillance
ISSN号	0162-8828
DOI	10.1109/TPAMI.2021.3084613
通讯作者	Zheng, Wei-Shi(wszheng@ieee.org)
英文摘要	In conventional person re-identification (re-id), the images used for model training in the training probe set and training gallery set are all assumed to be instance-level samples that are manually labeled from raw surveillance video (likely with the assistance of detection) in a frame-by-frame manner. This labeling across multiple non-overlapping camera views from raw video surveillance is expensive and time consuming. To overcome these issues, we consider a weakly supervised person re-id modeling that aims to find the raw video clips where a given target person appears. In our weakly supervised setting, during training, given a sample of a person captured in one camera view, our weakly supervised approach aims to train a re-id model without further instance-level labeling for this person in another camera view. The weak setting refers to matching a target person with an untrimmed gallery video where we only know that the identity appears in the video without the requirement of annotating the identity in any frame of the video during the training procedure. The weakly supervised person re-id is challenging since it not only suffers from the difficulties occurring in conventional person re-id (e.g., visual ambiguity and appearance variations caused by occlusions, pose variations, background clutter, etc.), but more importantly, is also challenged by weakly supervised information because the instance-level labels and the ground-truth locations for person instances (i.e., the ground-truth bounding boxes of person instances) are absent. To solve the weakly supervised person re-id problem, we develop deep graph metric learning (DGML). On the one hand, DGML measures the consistency between intra-video spatial graphs of consecutive frames, where the spatial graph captures neighborhood relationship about the detected person instances in each frame. On the other hand, DGML distinguishes the inter-video spatial graphs captured from different camera views at different sites simultaneously. To further explicitly embed weak supervision into the DGML and solve the weakly supervised person re-id problem, we introduce weakly supervised regularization (WSR), which utilizes multiple weak video-level labels to learn discriminative features by means of a weak identity loss and a cross-video alignment loss. We conduct extensive experiments to demonstrate the feasibility of the weakly supervised person re-id approach and its special cases (e.g., its bag-to-bag extension) and show that the proposed DGML is effective.
资助项目	National Key Research andDevelopment Program of China[2016YFB1001002] ; NSFC[U1911401] ; NSFC[U1811461] ; NSFC[U1803120] ; Guangdong NSF Project[2020B1515120085] ; Guangdong NSF Project[2018B030312002] ; Research Projects of Zhejiang Lab[2019KD0AB03] ; Key-Area Research and Development Program of Guangzhou[202007030004] ; Guangzhou Research Project[201902010037]
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE COMPUTER SOC
WOS记录号	WOS:000853875300019
资助机构	National Key Research andDevelopment Program of China ; NSFC ; Guangdong NSF Project ; Research Projects of Zhejiang Lab ; Key-Area Research and Development Program of Guangzhou ; Guangzhou Research Project
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/50119]
专题	自动化研究所_智能感知与计算研究中心
通讯作者	Zheng, Wei-Shi
作者单位	1.Sun Yat Sen Univ, Sch Comp Sci & Engn, Guangzhou 519082, Peoples R China 2.Pazhou Lab, Guangzhou 519082, Peoples R China 3.Sun Yat Sen Univ, Sch Comp Sci & Engn, Key Lab Machine Intelligence & Adv Comp, Minist Educ, Guangzhou 519082, Peoples R China 4.Chinese Acad Sci, Inst Automat, Beijing 100049, Peoples R China 5.Peng Cheng Lab, Shenzhen 518066, Peoples R China
推荐引用方式 GB/T 7714	Meng, Jingke,Zheng, Wei-Shi,Lai, Jian-Huang,et al. Deep Graph Metric Learning for Weakly Supervised Person Re-Identification[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2022,44(10):6074-6093.
APA	Meng, Jingke,Zheng, Wei-Shi,Lai, Jian-Huang,&Wang, Liang.(2022).Deep Graph Metric Learning for Weakly Supervised Person Re-Identification.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,44(10),6074-6093.
MLA	Meng, Jingke,et al."Deep Graph Metric Learning for Weakly Supervised Person Re-Identification".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 44.10(2022):6074-6093.