Robust pedestrian tracking using a model-based approach

This paper presents a real-time system for pedestrian tracking in sequences of grayscale images acquired by a stationary CCD camera. The objective is to integrate this system with a pedestrian control scheme for intersections. The system outputs the spatio-temporal coordinates of each pedestrian during the period the pedestrian is in the scene. Processing is done at three levels: raw images, blobs, and pedestrians. Our method models pedestrians as rectangular patches with a certain dynamic behavior. Kalman filtering is used to estimate pedestrian parameters. The system was implemented on a Datacube MaxVideo 20 equipped with a Datacube Max860 and was able to achieve a peak performance of over 20 frames per second. Experimental results based on indoor and outdoor scenes demonstrated the system's robustness under many difficult situations such as partial and full occlusions of pedestrians.