Centralized embedding hypersphere feature learning for person re-identification

ABSTRACT

Deep metric learning has become a general method for person re-identification (ReID) recently. Existing methods train ReID model with various loss functions to learn feature representation and identify pedestrian. However, the interaction between person features and classification vectors in the training process is rarely concerned. Distribution of pedestrian features will greatly affect convergence of the model and the pedestrian similarity computing in the test phase. In this paper, we formulate improved softmax function to learn pedestrian features and classification vectors. Our method applies pedestrian feature representation to be scattered across the coordinate space and embedding hypersphere to solve the classification problem. Then, we propose an end-to-end convolutional neural network (CNN) framework with improved softmax function to improve the performance of pedestrian features. Finally, experiments are performed on four challenging datasets. The results demonstrate that our work is competitive compared to the state-of-the-art.

KEYWORDS:

• Person re-identification
• embedding hypersphere
• centralized coordinate learning
• deep metric learning
• softmax function
• convolutional neural network
• deep learning
• embedding hypersphere

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Yuanyuan Wang received the M.S. degree in computer technology from the Nanjing University of Science and Technology in 2010. She is currently pursuing the Ph.D. degree in computer science and technology from the College of Computer and Information, Hohai University, China. She is currently a Lecturer with the College of Computer and Software Engineering, Huaiyin Institute of Technology, China. Her current research focuses on computer vision and person re-identification.

Zhijian Wang received the M.S. and Ph.D. degrees in computer science from Nanjing University, China. He is currently a Professor with the College of Computer and Information, Hohai University, China. His research interests include machine learning and computer application.

Mingxin Jiang received the Ph.D. degree in signal and information processing from the Dalian University of Technology, China, in 2013. She was a Post-Doctoral Researcher with the Department of Electrical Engineering, Dalian University of Technology. She is currently an Associate Professor with the College of Electronic Information Engineering, Huaiyin Institute of Technology. Her research interests include multi-object tracking and vision sensors for robotics.

ORCID

Yuanyuan Wang http://orcid.org/0000-0001-5152-5825

Additional information

Funding

The authors acknowledge the Natural Science Research of Jiangsu Higher Education Institutions of China [grant number 18KJA520002], the Natural Science Foundation of Jiangsu Province [grant number BK20171267], the Fifth Issue 333 High-Level Talent Training Project of Jiangsu Province [grant number BRA2018333].