模式识别与人工智能
Friday, Apr. 4, 2025 Home      About Journal      Editorial Board      Instructions      Ethics Statement      Contact Us                   中文
  2021, Vol. 34 Issue (9): 798-808    DOI: 10.16451/j.cnki.issn1003-6059.202109003
Deep Learning Algorithms and the Applications in Image and Vision Current Issue| Next Issue| Archive| Adv Search |
Person Re-identification Based on Fusion Relationship Learning Network
WU Ziqiang1, CHANG Hong1,2, MA Bingpeng1
1. School of Computer Science and Technology, University of Chinese Academy of Sciences, Beijing 100049
2. Key Laboratory of Intelligent Information Processing, Institute of Computing Technology, Chinese Academy of Sciences, Beijing 100190

Download: PDF (1007 KB)   HTML (1 KB) 
Export: BibTeX | EndNote (RIS)      
Abstract  There are two problems in person re-identification methods based on graph convolutional network(GCN). While graphs are built for feature maps, the semantic information represented by graph node is not salient. The process of selecting feature blocks to build graph just relies on the relative distance among feature blocks, and the content similarity is ignored. To settle these two problems, an algorithm of person re-identification based on fusion relationship learning network(FRLN) is proposed in this paper. By employing attention mechanism, the maximum attention model makes the most important feature block more salient and assigns semantic information to it. The affinities of feature blocks are evaluated by the fusion similarity metric in the aspect of distance and content, and thus the metric is more comprehensive. By the proposed algorithm, the neighbor feature blocks are selected comprehensively and better input graph structures are provided for GCN. Hence, more robust structure relationship features are extracted by GCN. Experiments on iLIDS-VID and MARS datasets verify the effectiveness of FRLN.
Key wordsPerson Re-identification      Graph Convolutional Network      Structural Relationship      Attention Mechanism      Similarity Metric Citation     
Received: 31 May 2021     
ZTFLH: TP 391.41  
Fund:National Natural Science Foundation of China(No.61876171,61976203), Open Project Fund from Shenzhen Institute of Artificial Intelligence and Robotics for Society (No.AC01
202005015)
Corresponding Authors: MA Bingpeng, Ph.D., associate professor. His research interests include principles and applications of pattern recognition, computer vision, artificial intelligence and machine learning.   
About author:: WU Ziqiang, master student. His research interests include person re-identification and natural language-based person search.
CHANG Hong, Ph.D., professor. Her research interests include algorithms and models in machine learning and pattern recognition, and their applications in image processing, computer vision and data mining.
Service
E-mail this article
Add to my bookshelf
Add to citation manager
E-mail Alert
RSS
Articles by authors
WU Ziqiang
CHANG Hong
MA Bingpeng
Cite this article:   
WU Ziqiang,CHANG Hong,MA Bingpeng. Person Re-identification Based on Fusion Relationship Learning Network[J]. , 2021, 34(9): 798-808.
URL:  
http://manu46.magtech.com.cn/Jweb_prai/EN/10.16451/j.cnki.issn1003-6059.202109003      OR     http://manu46.magtech.com.cn/Jweb_prai/EN/Y2021/V34/I9/798
Copyright © 2010 Editorial Office of Pattern Recognition and Artificial Intelligence
Address: No.350 Shushanhu Road, Hefei, Anhui Province, P.R. China Tel: 0551-65591176 Fax:0551-65591176 Email: bjb@iim.ac.cn
Supported by Beijing Magtech  Email:support@magtech.com.cn