Gait Identification Considering Body Tilt by Walking Direction Changes

ELCVIA Electronic Letters on Computer Vision and Image Analysis

Ryusuke Sagawa¹  Yasuhiro Mukaigawa¹  Yasushi Makihara¹  Yasushi Yagi¹  Tomio Echigo¹ 

关键词: Gait identification;    Body tilt correction;    Fourier analysis;    View Transformation Model;   

DOI  :  

学科分类：计算机科学（综合）

来源: ELCVIA

【 摘 要 】

Gait identification has recently gained attention as a method of identifying individuals at a distance.Thought most of the previous works mainly treated straight-walk sequences for simplicity, curved-walk sequencesshould be also treated considering situations where a person walks along a curved path or enters abuilding from a sidewalk. In such cases, person’s body sometimes tilts by centrifugal force when walkingdirections change, and this body tilt considerably degrades gait silhouette and identification performance,especially for widely-used appearance-based approaches. Therefore, we propose a method of body-tilted silhouettecorrection based on centrifugal force estimation from walking trajectories. Then, gait identificationprocess including gait feature extraction in the frequency domain and learning of a View TransformationModel (VTM) follows the silhouette correction. Experiments of gait identification for circular-walk sequencesdemonstrate the effectiveness of the proposed method.

【 授权许可】

Unknown