View-Invariant Gait Recognition Using Gait Energy Image (GEI)

Biometric systems are becoming increasingly important, as they provide more reliable and efficient means of identity verification. Recognition of a person from their gait is a biometric of increasing interest among computer vision researchers. Biometric system is essentially a gait which recognizes people based on the way they walk. In our paper works a new innovative proposal to recognize gait for human identification using Gaussian mixture model. First the background modelling is done from a video sequence. Afterwards, the moving foreground objects in the individual image frames are segmented using the background subtraction algorithm. Eventually, the skeleton is used to track the moving silhouettes of a walking figure. Finally, when a video sequence is fed, the recognized the gait features and thereby peoples based on self-similarity measure. This paper gives a synopsis of recognition of gait obtained from faraway distance and also dealing with low resolution videos. Gait recognition is one kind of biometric technique that can be used to monitor with their partisanship. Gait recognition not only used for video surveillance, also medical diagnostic, biometric identification and forensics, comparative biomechanics and sports environments.