Uncertainty-optimized deep learning model for small-scale person re-identification

doi:10.1007/s11432-019-2675-3

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

	Uncertainty-optimized deep learning model for small-scale person re-identification
	Zhao, Cairong 2; Chen, Kang 2; Zang, Di 2; Zhang, Zhaoxiang 3; Zuo, Wangmeng 1; Mia, Duoqian 2
刊名	SCIENCE CHINA-INFORMATION SCIENCES
	2019-12-01
卷号	62 期号:12 页码:13
关键词	person re-identification uncertainty analysis deep learning
ISSN号	1674-733X
DOI	10.1007/s11432-019-2675-3
通讯作者	Zhao, Cairong(zhaocairong@tongji.edu.cn)
英文摘要	In recent years, deep learning has developed rapidly and is widely used in various fields, such as computer vision, speech recognition, and natural language processing. For end-to-end person re-identification, most deep learning methods rely on large-scale datasets. Relatively few methods work with small-scale datasets. Insufficient training samples will affect neural network accuracy significantly. This problem limits the practical application of person re-identification. For small-scale person re-identification, the uncertainty of person representation and the overfitting problem associated with deep learning remain to be solved. Quantifying the uncertainty is difficult owing to complex network structures and the large number of hyperparameters. In this study, we consider the uncertainty of pedestrian representation for small-scale person re-identification. To reduce the impact of uncertain person representations, we transform parameters into distributions and conduct multiple sampling by using multilevel dropout in a testing process. We design an improved Monte Carlo strategy that considers both the average distance and shortest distance for matching and ranking. When compared with state-of-the-art methods, the proposed method significantly improve accuracy on two small-scale person re-identification datasets and is robust on four large-scale datasets.
资助项目	National Natural Science Foundation of China[61673299] ; National Natural Science Foundation of China[61203247] ; National Natural Science Foundation of China[61573259] ; National Natural Science Foundation of China[61573255] ; National Natural Science Foundation of China[61876218] ; Fundamental Research Funds for the Central Universities ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR)
WOS关键词	GAP
WOS研究方向	Computer Science ; Engineering
语种	英语
出版者	SCIENCE PRESS
WOS记录号	WOS:000498592200001
资助机构	National Natural Science Foundation of China ; Fundamental Research Funds for the Central Universities ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR)
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/29355]
专题	自动化研究所_智能感知与计算研究中心
通讯作者	Zhao, Cairong
作者单位	1.Harbin Inst Technol, Sch Comp Sci & Technol, Harbin 150001, Heilongjiang, Peoples R China 2.Tongji Univ, Dept Comp Sci & Technol, Shanghai 201804, Peoples R China 3.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Zhao, Cairong,Chen, Kang,Zang, Di,et al. Uncertainty-optimized deep learning model for small-scale person re-identification[J]. SCIENCE CHINA-INFORMATION SCIENCES,2019,62(12):13.
APA	Zhao, Cairong,Chen, Kang,Zang, Di,Zhang, Zhaoxiang,Zuo, Wangmeng,&Mia, Duoqian.(2019).Uncertainty-optimized deep learning model for small-scale person re-identification.SCIENCE CHINA-INFORMATION SCIENCES,62(12),13.
MLA	Zhao, Cairong,et al."Uncertainty-optimized deep learning model for small-scale person re-identification".SCIENCE CHINA-INFORMATION SCIENCES 62.12(2019):13.