International Journal of Advanced Research in Computer Science and Software Engineering (IJARCSSE)
Human recognition is always been an important task for various scenario like wide area monitoring, asset protection, security etc. Here a new dataset for gait recognition is presented. The presented database overcomes a crucial limitation of other gait recognition databases. This dataset offers four new kinds of gait variations include hands in pocket, wearing backpack and wearing a gown which allow for challenging evaluation of recognition algorithms. In addition to presenting the database the authors present algorithm like gait energy image and PCA to perform person identification using gait.