Exploiting temporal statistics for events analysis and understanding

In this paper, we propose a technique for detecting anomalous events in outdoor areas monitored by a video surveillance system. In particular, the focus is on the time spent by an object to carry out simple events. To have a statistical representation of the time commonly required to perform certain activities, mixtures of Gaussians are maintained for each event type. Such statistics are then exploited both for the analysis of simple activities and for discovering anomalous situations, eventually alerting the operator. To this end, a novel way of visualizing results is also discussed. Experiments have been performed on a multi-camera system for parking lot security.