Person Re-identification Based on Attribute Hierarchy Recognition

被引：3

作者：

Chen Hongchang ^{[1
]}

Wu Yancheng ^{[1
]}

Li Shaomei ^{[1
]}

Gao Chao ^{[1
]}

机构：

[1] China Natl Digital Switching Syst Engn & Technol, Zhengzhou 450002, Henan, Peoples R China

来源：

JOURNAL OF ELECTRONICS & INFORMATION TECHNOLOGY | 2019年 / 41卷 / 09期

基金：

中国国家自然科学基金;

关键词：

Person re-identification; Attention model; Deep learning; Saliency; Hierarchy; NETWORK;

D O I：

10.11999/JEIT180740

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

In order to improve the accuracy rate of person re-identification, a pedestrian attribute hierarchy recognition neural network is proposed based on attention model. Compared with the existing algorithms, the model has the following three advantages. Firstly, the attention model is used in this paper to identify the pedestrian attributes, and to extract of pedestrian attribute information and degree of significance. Secondly, the attention model in used in this paper to classify the attributes according to the significance of the pedestrian attributes and the amount of information contained. Thirdly, this paper analyzes the correlation between attributes, and adjusts the next level identification strategy according to the recognition results of the upper level. It can improve the recognition accuracy of small target attributes, and the accuracy of pedestrian recognition is improved. The experimental results show that the proposed model can effectively improve the first accuracy rate (rank-1) of person re-identification compared with the existing methods. On the Market1501 dataset, the first accuracy rate is 93.1%, and the first accuracy rate is 81.7% on the DukeMTMC dataset.

引用

页码：2239 / 2246

页数：8

共 20 条

[1] Similarity Learning with Spatial Constraints for Person Re-identification [J].

Chen, Dapeng ;

Yuan, Zejian ;

Chen, Badong ;

Zheng, Nanning .

2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :1268-1277

[2] Person Re-identification of Adaptive Blocks Based on Saliency Fusion [J].

Chen, Hongchang ;

Chen, Lei ;

Li, Shaomei ;

Zhu, Junguang .

Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology, 2017, 39 (11) :2652-2660

[3] Person Re-Identification by Symmetry-Driven Accumulation of Local Features [J].

Farenzena, M. ;

Bazzani, L. ;

Perina, A. ;

Murino, V. ;

Cristani, M. .

2010 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2010, :2360-2367

[4]

He K., 2016, CVPR, DOI [10.1109/CVPR.2016.90, DOI 10.1109/CVPR.2016.90]

[5]

KOSTINGER M, 2012, PROC CVPR IEEE, P2288, DOI DOI 10.1109/CVPR.2012.6247939

[6] Learning Deep Context-aware Features over Body and Latent Parts for Person Re-identification [J].

Li, Dangwei ;

Chen, Xiaotang ;

Zhang, Zhang ;

Huang, Kaiqi .

30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017), 2017, :7398-7407

[7] Harmonious Attention Network for Person Re-Identification [J].

Li, Wei ;

Zhu, Xiatian ;

Gong, Shaogang .

2018 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2018, :2285-2294

[8]

Li W, 2017, PROCEEDINGS OF THE TWENTY-SIXTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, P2194

[9] DeepReID: Deep Filter Pairing Neural Network for Person Re-Identification [J].

Li, Wei ;

Zhao, Rui ;

Xiao, Tong ;

Wang, Xiaogang .

2014 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2014, :152-159

[10]

Liao SC, 2015, PROC CVPR IEEE, P2197, DOI 10.1109/CVPR.2015.7298832

← 1 2 →