Earticle

초록

영어

Automatic gender classification of an individual can be very useful in video-based surveillance systems and human-computer interaction systems. Currently, gait from a single viewpoint has been used to recognize the gender of a person. Considering the multiple cameras used in real environments, we investigate gender classification from human gait by using multi-view fusion, a relatively understudied problem. In this paper, we present a new approach to integrate information from multi-view gait at the feature level. First, gait energy images (GEI) are constructed from the video streams for different viewpoints. Then, the feature fusion is performed by putting GEI images and camera views together to generate a third-order tensor (x, y, view). A multi-linear principal component analysis (MPCA) is employed to reduce dimensionality of the tensor objects which integrate all views. The proposed fusion scheme is tested on CASIA database and compared with other fusion methods. The experimental results show that MPCA based feature fusion is quite effective for multi-view gait based gender classification.

목차

Abstract
 1. Introduction
 2. Related Work
 3. Technical Approach
  3.1. GEI Construction
  3.2. Fusing Multi-View Gait
  3.3. Feature Learning Using MPCA
  3.4. Related Fusion Schemes
 4. Experiments
  4.1. Database
  4.2. Experimental Results
  4.3. Comparison with Other Related Work
  4.4. Discussion
 5. Conclusion
 Acknowledgements
 References

키워드

저자

• Zhang De [ School of Electrical and Information Engineering, Beijing University of Civil Engineering and Architecture, Beijing, 100044, P.R. China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Database Theory and Application

• 간기

  격월간

• pISSN

  2005-4270

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605