Partial Person Re-identification

被引:244
作者
Zheng, Wei-Shi [1 ]
Li, Xiang [1 ]
Xiang, Tao [2 ]
Liao, Shengcai [3 ]
Lai, Jianhuang [1 ]
Gong, Shaogang [2 ]
机构
[1] Sun Yat Sen Univ, Sch Informat Sci & Technol, Guangzhou, Guangdong, Peoples R China
[2] Queen Mary Univ London, Sch Elect Engn & Comp Sci, London, England
[3] Chinese Acad Sci, Inst Automat, NLPR, Beijing 100864, Peoples R China
来源
2015 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV) | 2015年
关键词
D O I
10.1109/ICCV.2015.531
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We address a new partial person re-identification (reid) problem, where only a partial observation of a person is available for matching across different non-overlapping camera views. This differs significantly from the conventional person re-id setting where it is assumed that the full body of a person is detected and aligned. To solve this more challenging and realistic re-id problem without the implicit assumption of manual body-parts alignment, we propose a matching framework consisting of 1) a local patch-level matching model based on a novel sparse representation classification formulation with explicit patch ambiguity modelling, and 2) a global part-based matching model providing complementary spatial layout information. Our framework is evaluated on a new partial person re-id dataset as well as two existing datasets modified to include partial person images. The results show that the proposed method outperforms significantly existing re-id methods as well as other partial visual matching methods.
引用
收藏
页码:4678 / 4686
页数:9
相关论文
共 43 条
[1]  
Andres A. M., 2014, PRL, V36
[2]  
[Anonymous], 2013, CVPR
[3]  
[Anonymous], MULTIMEDIA MODELING
[4]  
[Anonymous], 2007, ICCV
[5]  
[Anonymous], 2011, CVPR
[6]  
[Anonymous], 2010, BMVC
[7]  
[Anonymous], 2013, ICCV
[8]  
[Anonymous], IEEE T PAMI
[9]  
[Anonymous], 2013, CVPR
[10]  
[Anonymous], 2014, ECCV