Biometric systems are becoming increasingly important since they provide more reliable and efficient means of identity verification. Identification of a person based on gait has created a sphere of curiosity in the computer vision domain due to its high recognition capability even at a far distance with simple instrumentation. It does not require the cooperation or even awareness of the individual under observation. Gait recognition technology has many civilian and highsecurity applications like car parks, banks, military bases, railway stations and airports. The project aims to develop an automatic biometric system to identify a person based on his Gait cycle. We can achieve this by converting the live video into still frames and applying feature extracting techniques to get silhouettes from where we can differentiate different gait cycles. These gait cycles can then be compared to the already existing data in the database to identify the subject.

Keywords : Identification, Cooperation, Security, Gait Cycle, Silhouettes, Database