Image to Video Person Re-Identification by Learning Heterogeneous Dictionary Pair With Feature Projection Matrix

doi:10.1109/TIFS.2017.2765524

CORC > 计算技术研究所 > 中国科学院计算技术研究所 > 中国科学院计算技术研究所期刊论文 > 英文

	Image to Video Person Re-Identification by Learning Heterogeneous Dictionary Pair With Feature Projection Matrix
	Zhu, Xiaoke 7,8; Jing, Xiao-Yuan 6,8; You, Xinge 5; Zuo, Wangmeng 4; Shan, Shiguang 2,3; Zheng, Wei-Shi 1
刊名	IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
	2018-03-01
卷号	13 期号:3 页码:717-732
关键词	Person re-identification image to video person re-identification heterogeneous dictionary pair learning feature projection matrix multi-view learning
ISSN号	1556-6013
DOI	10.1109/TIFS.2017.2765524
英文摘要	Person re-identification plays an important role in video surveillance and forensics applications. In many cases, person re-identification needs to be conducted between image and video clip, e.g., re-identifying a suspect from large quantities of pedestrian videos given a single image of the suspect. We call re-identification in this scenario as image to video person re-identification (IVPR). In practice, image and video are usually represented with different features, and there usually exist large variations between frames within each video. These factors make matching between image and video become a very challenging task. In this paper, we propose a joint feature projection matrix and heterogeneous dictionary pair learning (PHDL) approach for IVPR. Specifically, the PHDL jointly learns an intra-video projection matrix and a pair of heterogeneous image and video dictionaries. With the learned projection matrix, the influence caused by the variations within each video on the matching can be reduced. With the learned dictionary pair, the heterogeneous image and video features can be transformed into coding coefficients with the same dimension, such that the matching can be conducted by using the coding coefficients. Furthermore, to ensure that the obtained coding coefficients own favorable discriminability, the PHDL designs a point-to-set coefficient discriminant term. To make better use of the complementary spatial-temporal and visual appearance information contained in pedestrian video data, we further propose a multi-view PHDL approach, which can fuse different video information effectively in the dictionary learning process. Experiments on four publicly available person sequence data sets demonstrate the effectiveness of the proposed approaches.
资助项目	National Key Research and Development Program of China[2017YFB0202001] ; National Nature Science Foundation of China[61671182] ; National Nature Science Foundation of China[61672208] ; National Nature Science Foundation of China[61772220] ; National Nature Science Foundation of China[41571417] ; National Nature Science Foundation of China[U1404618] ; National Nature Science Foundation of China[61272273] ; National Nature Science Foundation of China[61572375] ; National Nature Science Foundation of China[61233011] ; National Nature Science Foundation of China[91418202] ; National Nature Science Foundation of China[61472178] ; National Nature Science Foundation of China[61373038] ; National Nature Science Foundation of China[61672392] ; National Basic Research 973 Program of China[2014CB340702] ; Ministry of Science and Technology of China[2015BAK36B00] ; Ministry of Science and Technology of China[2015BAK01B06] ; Key Science and Technology of Shenzhen[CXZZ20150814155434903] ; Key Program for International S&T Cooperation Projects of China[2016YFE0121200] ; Natural Science Foundation of Jiangsu Province[BK20170900] ; Scientific Research Staring Foundation for Introduced Talents in NJUPT under NUPTSF[NY217009] ; Science and Technology Program in Henan Province[1721102410064] ; Science and Technology Program in Henan Province[172102210186] ; Research Foundation of Henan University[2015YBZR024]
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
WOS记录号	WOS:000418723000014
内容类型	期刊论文
源URL	[http://119.78.100.204/handle/2XEOYT63/6353]
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Jing, Xiao-Yuan
作者单位	1.Sun Yat Sen Univ, Sch Data & Comp Sci, Guangzhou, Guangdong, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Beijing, Peoples R China 3.Chinese Acad Sci, Key Lab Intelligent Informat Proc, Beijing, Peoples R China 4.Harbin Inst Technol, Sch Comp Sci & Technol, Harbin, Heilongjiang, Peoples R China 5.Huazhong Univ Sci & Technol, Sch Elect Informat & Commun, Wuhan, Hubei, Peoples R China 6.Nanjing Univ Posts & Telecommun, Sch Automat, Nanjing 210023, Jiangsu, Peoples R China 7.Henan Univ, Sch Comp & Informat Engn, Kaifeng 475001, Peoples R China 8.Wuhan Univ, Sch Comp, State Key Lab Software Engn, Wuhan 430072, Peoples R China
推荐引用方式 GB/T 7714	Zhu, Xiaoke,Jing, Xiao-Yuan,You, Xinge,et al. Image to Video Person Re-Identification by Learning Heterogeneous Dictionary Pair With Feature Projection Matrix[J]. IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY,2018,13(3):717-732.
APA	Zhu, Xiaoke,Jing, Xiao-Yuan,You, Xinge,Zuo, Wangmeng,Shan, Shiguang,&Zheng, Wei-Shi.(2018).Image to Video Person Re-Identification by Learning Heterogeneous Dictionary Pair With Feature Projection Matrix.IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY,13(3),717-732.
MLA	Zhu, Xiaoke,et al."Image to Video Person Re-Identification by Learning Heterogeneous Dictionary Pair With Feature Projection Matrix".IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 13.3(2018):717-732.