76
views
0
recommends
+1 Recommend
1 collections
    2
    shares
      • Record: found
      • Abstract: found
      • Article: found

      Improved Bag-of-Words Model for Person Re-identification

      research-article

      Read this article at

      ScienceOpenPublisher
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Person re-identification (person re-id) aims to match observations on pedestrians from different cameras. It is a challenging task in real word surveillance systems and draws extensive attention from the community. Most existing methods are based on supervised learning which requires a large number of labeled data. In this paper, we develop a robust unsupervised learning approach for person re-id. We propose an improved Bag-of-Words (iBoW) model to describe and match pedestrians under different camera views. The proposed descriptor does not require any re-id labels, and is robust against pedestrian variations. Experiments show the proposed iBoW descriptor outperforms other unsupervised methods. By combination with efficient metric learning algorithms, we obtained competitive accuracy compared to existing state-of-the-art methods on person re-identification benchmarks, including VIPeR, PRID450S, and Market1501.

          Author and article information

          Journal
          Tsinghua Science and Technology
          Tsinghua Science and Technology
          Tsinghua University Press (Xueyan Building, Tsinghua University, Beijing 100084, China )
          1007-0214
          05 April 2018
          : 23
          : 2
          : 145-156 (pp. )
          Affiliations
          [1]∙ Lu Tian and Shengjin Wang are with the Department of Electronic Engineering, Tsinghua University, Beijing 100084, China.
          Author notes
          * To whom correspondence should be addressed. E-mail: wgsgj@ 123456tsinghua.edu.cn .

          Shengjin Wang  received the BEng degree from Tsinghua University, China, and the PhD degree from the Tokyo Institute of Technology, Tokyo, Japan, in 1985 and 1997, respectively. From 1997 to 2003, he was a member of the researcher with the Internet System Research Laboratories, NEC Corporation, Japan. Since 2003, he has been a professor with the Department of Electronic Engineering, Tsinghua University, where he is currently the director of the Research Institute of Image and Graphics. His current research interests include image processing, computer vision, video surveillance, and pattern recognition. He is a member of IEEE and IEICE.

          Lu Tian received the BEng degree from Tsinghua University, China, in 2011. She has been studying for the PhD degree in electronic engineering at Tsinghua University from 2011. Her current research interests include pattern recognition and human feature extraction, in particular person re-identification.

          Article
          1007-0214-23-2-145
          10.26599/TST.2018.9010060
          eb417c3e-67d8-4743-b9fb-43b28c616d34
          Copyright @ 2018
          History
          : 04 December 2016
          : 25 January 2017
          : 22 January 2017
          Categories
          Regular Article

          Software engineering,Data structures & Algorithms,Applied computer science,Computer science,Artificial intelligence,Hardware architecture
          unsupervised learning,bag-of-words,person re-identification,feature fusion

          Comments

          Comment on this article