PROJECT TITLE :

Information fusion from multiple cameras for gait-based re-identification and recognition

ABSTRACT:

During this study, the authors gift a absolutely automated frontal (i.e. employing front and back views only) gait recognition approach using the depth information captured by multiple Kinect RGB-D cameras. Restricted depth sensing range restricts every of these Kinects to record solely a part of a whole gait cycle of a walking subject. Hence, data from a lot of than one Kinect is fused along to examine that features of a gait cycle can be conveniently extracted from the sequences captured independently by these cameras. To achieve this, it is imperative that the same subject be re-identified as he moves from the field of read of 1 camera to a different. The authors use a set of soft-biometric options computed from the skeleton stream provided by Kinect software development kit) for doing automatic re-identification. To enable such information fusion and conjointly to handle missing components even after re-identification, features are extracted at the granularity of tiny fractions of a gait cycle. Experiments applied on a knowledge set with gait videos captured by Kinects respectively from the back and front views show promising results.