Face tracking using Kalman filter with dynamic noise statistics

We present a system for automatic detection and tracking of faces in video sequences. Detection is done based on a statistical characterization of skin-color. The position and size of the dominant face are estimated using statistical means of the binary map projections. Tracking is done using a Kalman filter. We propose a novel technique for updating the process noise covariance at each iteration. The algorithm performs well on a variety of real-life videos. The algorithm is (1) able to automatically detect the initial position and size of face, (2) relatively insensitive to lighting condition variations, (3) robust against partial occlusions (4) able to track in case of scale changes. Experimental results demonstrate the effectiveness of the algorithm on a variety of videos.