Adaptation to Walking Direction Changes for Gait Identification

This paper describes adaptation to gradual changes of walking directions for gait identification. First, we propose a method of body tilt correction due to changes of walking directions when constructing a spatio-temporal gait silhouette volume. Next, we propose a view transformation model in the frequency domain to match gait features of different walking directions. Finally, experiments of gait identification for a circular path demonstrate the effectiveness of the proposed method.

• Type: Conference
• Year: 2006
• Where: ICPR
• Authors: Ryusuke Sagawa, Tomio Echigo, Yasuhiro Mukaigawa, Yasushi Makihara, Yasushi Yagi
• Keywords: Computer Vision, Gait Features, Gait Identification, ICPR 2006, Spatio-temporal Gait Silhouette