NEUROCOMPUTATIONAL BASES OF OBJECT AND FACE RECOGNITION

A number of behavioural phenomena distinguish the recognition of faces and objects, even when members of a set of objects are highly similar . Because faces have the same parts in approximately the same relation s, individuation of faces typically requires specification of the metr ic variation in a holistic and integral representation of the facial s urface. The direct mapping of a hypercolumn-like pattern of activation onto a representation layer that preserves relative spatial filter va lues in a two-dimensional (2D) coordinate space, as proposed by C. von der Malsburg and his associates, may account for many of the phenomen a associated with face recognition. An additional refinement, in which each column of filters (termed at 'jet') is centred on a particular f acial feature (or fiducial point), allows selectivity of the input int o the holistic representation to avoid incorporation of occluding or n earby surfaces. The initial hypercolumn representation also characteri zes the first stage of object perception, but the image variation for objects at a given location in a 2D coordinate space may be too great to yield sufficient predictability directly from the output of spatial kernels. Consequently, objects can be represented by a structural des cription specifying qualitative (typically, non-accidental) characteri zations of an object's parts, the attributes of the parts, and the rel ations among the parts, largely based on orientation and depth discont inuities (as shown by Hummel & Biederman). A series of experiments on the name priming or physical matching of complementary images (in the Fourier domain) of objects and faces documents that whereas face recog nition is strongly dependent on the original spatial filter values, ev idence from object recognition indicates strong invariance to these va lues, even when distinguishing among objects that are as similar as fa ces.