Automatic Gait Recognition via Model-Based Evidence Gathering

Download

PDF Download (60Kb)

Description/Abstract

Recognising people by gait is of emergent interest. A new model-based moving feature extraction analysis is presented that automatically extracts and describes human gait for recognition. The gait signature is extracted directly from the evidence gathering process by using a Fourier series to describe the motion of the upper leg and apply temporal evidence gathering techniques to extract the moving model from a sequence of images. Performance has been confirmed by simulation and on a small subject database and verified using statistical analysis of the separation of clusters in feature space. Further, the technique can handle occlusion, of especial importance in gait as the human body is self-occluding. As such, a new technique has been developed to automatically extract and describe a moving articulated shape, the human leg, and shown its potential in gait as a biometric.