DESCRIBING THE SHAPES OF FACES USING SURFACE PRIMITIVES

Humans are poor at recognizing faces from images that convey the spatial layout of major features but that omit information (e.g. simple line drawings) or invert information (photographic negatives) about the image intensity. The role played by the image intensity may be to convey useful information about surface pigmentation an&or 3D shape. To investigate the role played by the latter, in this paper we describe the development of a method for describing the 3D surface shape of individual faces explicitly, and consider some initial results which suggest that this method does capture some psychologically relevant dimensions of facial variation. Implications for machine recognition and other practical computer applications are discussed.