Multiple Views Effective for Gait Recognition Based on Contours

Abstract

Gait is one of well recognized biometrics that has been widely used for human identification. However, the current gait recognition might have difficulties due to viewing angle being changed. This is because the viewing angle under which the gait signature database was generated may not be the same as the viewing angle when the probe data are obtained. This paper present an effective multi-view gait recognition based on motion contour (MVGRMC) approach which tackles the problems mentioned above.  Initially the background modeling is done from a video sequence. Subsequently, the moving foreground objects in the individual image frames are segmented using the background subtraction algorithm. Then, the morphological skeleton operator is used to track the moving silhouettes of a walking , Finally, when a video sequence is fed, the proposed system recognizes the gait features and thereby humans, based on self-similarity measure. The proposed system is evaluated using gait databases and the experimentation on outdoor video sequences demonstrates that the proposed algorithm achieves a pleasing recognition performance. The proposed algorithm can significantly improve the multiple view gait recognition performance when being compared to the similar methods. These results are illustrated with practical examples on popular gait databases.

Keywords: Gait Recognition; Biometric; silhouette; Motion analysis; Feature extraction