ZTE Communications ›› 2021, Vol. 19 ›› Issue (1): 72-81.DOI: 10.12142/ZTECOM.202101009

• Research Paper • Previous Articles     Next Articles

Integrating Coarse Granularity Part-Level Features with Supervised Global-Level Features for Person Re-Identification

CAO Jiahao1,2(), MAO Xiaofei1,2, LI Dongfang1,2, ZHENG Qingfang1,2, JIA Xia1,2   

  1. 1.State Key Laboratory of Mobile Network and Mobile Multimedia Technology, Shenzhen 518057, China
    2.ZTE Corporation, Shenzhen 518057, China
  • Online:2021-03-25 Published:2021-04-09
  • About author:CAO Jiahao (cao.jiahao@zte.com.cn) received the M.S. degree from Northeastern University, China in 2019 and joined ZTE corporation after he graduated. His current research interests include image processing and deep learning technologies.|MAO Xiaofei received the M.S. degree from TELECOM ParisTech, France in 2017. His current research interests include person re-identification, image processing and deep learning technologies.|LI Dongfang received the M.S. degree in electronics and communications engineering from Harbin Engineering University, China in 2017. He has been engaged in deep learning technologies in ZTE Corporation since his graduation.|ZHENG Qingfang received the B.S. degree in civil engineering and computer applications from Shanghai Jiaotong University, China in 2002, and the Ph.D. degree in computer sciences from Chinese Academy of Sciences, China in 2008. He is currently the chief scientist of video technology in ZTE Corporation. His research interests include computer vision, video codec, video streaming and multimedia content analysis and retrieval. He has published around 10 papers in various journals and conferences.|JIA Xia received her B.S. degree and M.S. degree in control theory and control engineering from Taiyuan University of Technology and Dalian University of Technology, China in 1995 and 2001, respectively. She joined ZTE Corporation in 2001 and worked in the State Key Laboratory of Mobile Network and Mobile Multimedia Technology. Her main research interests include deep learning techniques, face detection and recognition, Re-ID, and activity detection and recognition.


Person re-identification (Re-ID) has achieved great progress in recent years. However, person Re-ID methods are still suffering from body part missing and occlusion problems, which makes the learned representations less reliable. In this paper, we propose a robust coarse granularity part-level network (CGPN) for person Re-ID, which extracts robust regional features and integrates supervised global features for pedestrian images. CGPN gains two-fold benefit toward higher accuracy for person Re-ID. On one hand, CGPN learns to extract effective regional features for pedestrian images. On the other hand, compared with extracting global features directly by backbone network, CGPN learns to extract more accurate global features with a supervision strategy. The single model trained on three Re-ID datasets achieves state-of-the-art performances. Especially on CUHK03, the most challenging Re-ID dataset, we obtain a top result of Rank-1/mean average precision (mAP)=87.1%/83.6% without re-ranking.

Key words: person Re-ID, supervision, coarse granularity