Real-time tracking of multiple people through stereo vision

People tracking is an important functionality for many applications that require to analyze the interaction between people and the environment. These applications range from security and environmental surveillance, to ambient intelligence. In this paper we describe a real-time People Localization and Tracking (PLT) System, based on a calibrated fixed stereo vision sensor. The system is able to locate and track multiple people in the environment, by using three different representations of the information coming from a stereo vision sensor: the left intensity image, the disparity image, and the 3-D world locations of measured points, as well as people modeling techniques in order to reacquire people that are temporary occluded during tracking. Experimental results show that the system is able to track multiple people moving in an area approximately 3 × 8 meters in front of the sensor with high reliability and good precision.