Behavior Understanding at Railway Station by Association of Locational Semantics and Postures

Abstract

This paper describes a framework of behavior understanding at railway station by association of location semantics and postures. The framework can interpret a time-series behavior of objects using results of the object tracking by the S-T MRF model. The proposed framework succeeds in understanding the abnormal behavior assumed at the station, and is shown the effectiveness as a result of experiments according to the image that imitates the behavior at the station.