Fast frontal view gait authentication based on the statistical registration and human gait modeling

We study the problem of analyzing and classifying frontal view gait video data. In this study, we suppose that frontal view gait data as a mixing of scale changing, human movements and speed changing parameters. We estimate these parameters using the statistical registration and modeling on a video data. To demonstrate the effectiveness of our method, we conducted experiment, assessing the proposed method for frontal view human gait authentication. We apply K-nearestneighbor classifier, using the estimated parameters, to perform the human gait authentication, and present results from an experiment involving 120 subjects. As a result, our method shows high recognition rate and low calculation cost.