DocumentCode :
3209965
Title :
Statistical feature fusion for gait-based human recognition
Author :
Han, Ju ; Bhanu, Bir
Author_Institution :
Center for Res. in Intelligent Syst., California Univ., Riverside, CA, USA
Volume :
2
fYear :
2004
fDate :
27 June-2 July 2004
Abstract :
This paper presents a novel approach for human recognition by combining statistical gait features from real and synthetic templates. Real templates are directly computed from training silhouette sequences, while synthetic templates are generated from training sequences by simulating silhouette distortion. A statistical feature extraction approach is used for learning effective features from real and synthetic templates. Features learned from real templates characterize human walking properties provided in training sequences, and features learned from synthetic templates predict gait properties under other conditions. A feature fusion strategy is therefore applied at the decision level to improve recognition performance. We apply the proposed approach to USF HumanID Database. Experimental results demonstrate that the proposed fusion approach not only achieves better performance than individual approaches, but also provides large improvement in performance with respect to the baseline algorithm.
Keywords :
database management systems; distortion; face recognition; feature extraction; image motion analysis; learning (artificial intelligence); USF HumanID Database; baseline algorithm; effective features learning; gait-based human recognition; human walking properties; real templates; silhouette distortion; statistical feature extraction approach; statistical feature fusion; statistical gait features; synthetic templates; training silhouette sequences; Biological system modeling; Biometrics; Character recognition; Fingerprint recognition; Frequency measurement; Humans; Intelligent systems; Legged locomotion; Pattern recognition; Phase measurement;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Vision and Pattern Recognition, 2004. CVPR 2004. Proceedings of the 2004 IEEE Computer Society Conference on
ISSN :
1063-6919
Print_ISBN :
0-7695-2158-4
Type :
conf
DOI :
10.1109/CVPR.2004.1315252
Filename :
1315252
Link To Document :
بازگشت