FACIAL FEATURE DETECTION USING GEOMETRICAL FACE MODEL - AN EFFICIENT APPROACH

A useful geometrical face model and an efficient facial feature detect ion approach are proposed. Based on the fact that human faces are cons tructed in the same geometrical configuration, the proposed approach c an accurately detect facial features, especially the eyes, even when t he images have complex backgrounds. The average computation time for o ne image of size 512 x 340 is less than 5 s by a SUN-Spare 20 workstat ion. Experimental results demonstrate that the proposed approach can e fficiently detect human facial features and satisfactorily deal with t he problems caused by bad lighting condition, skew face orientation, a nd even facial expression. (C) 1997 Pattern Recognition Society. Publi shed by Elsevier Science Ltd.