Automatic Face Recognition: What Representation?

We describe a testbed which can be used to investigate diierent codings for automatic face recognition. An eigenface coding of shape-free faces, based on manually coded landmarks was found to be more eeective than the corresponding coding of correctly shaped faces. The advantage for shape-free faces was found to reeect and depend upon high-quality representation of the facial variation by means of an ensemble of shape-free faces. Connguration also proved an eeective method of recognition , with rankings given to incorrect matches relatively uncorrelated with those from shape-free faces. Both sets of information combine to improve signiicantly the performance of either system. Manipulation within the coding to emphasize distinctive features of the faces, by caricaturing, allowed further increases in performance; this eeect was noticeably larger when the independent shape-free and connguration coding was used. The addition of a system which directly correlated the contours of shape-free images also signiicantly increased recognition, suggesting extra information was still available. The coding of the connguration provided a means of automatic measurement of the face-shape using an active shape model. Taken together , these results strongly support the suggestion that faces should be considered as lying in a high-dimensional manifold which is linearly approximated by these two factors, possibly with a separate system for local features. Abstract We describe a testbed which can be used to investigate diierent codings for automatic face recognition. An eigenface coding of shape-free faces, based on manually coded landmarks was found to be more eeective than the corresponding coding of correctly shaped faces. The advantage for shape-free faces was found to reeect and depend upon high-quality representation of the facial variation by means of an ensemble of shape-free faces. Connguration also proved an eeective method of recognition , with rankings given to incorrect matches relatively uncorrelated with those from shape-free faces. Both sets of information combine to improve signiicantly the performance of either system. Manipulation within the coding to emphasize distinctive features of the faces, by caricaturing, allowed further increases in performance; this eeect was noticeably larger when the independent shape-free and connguration coding was used. The addition of a system which directly correlated the contours of shape-free images also signiicantly increased recognition, suggesting extra information was still available. The coding of the connguration provided a means of automatic measurement of the face-shape using an active shape model. Taken together , these results strongly support the suggestion that faces …