Automatic location tracking of faces and facial features in video sequences

The work reported in this paper addresses the issue of automatically tracking the faces and facial features of persons in head-and-shoulders video sequences. We propose two totally automatic algorithms which respectively perform the detection of head outlines and identify rectangular \eyes-nose-mouth" regions, both from downsampled binary thresholded edge images. Unlike ones that have been proposed recently, a priori assumptions regarding the nature and content of the sequences to code are minimal for our techniques, and the algorithms operate accurately and robustly, even in cases of signi cant head rotation or partial occlusion by moving objects.