CARICATURAL EFFECTS IN AUTOMATED FACE PERCEPTION

This paper analyzes properties of a certain class of approximation tec hniques - HyperBF networks - in face perception tasks. The problem of gender classification and identification is addressed using a geometri cal description of faces, extracted automatically from digitized pictu res of frontal views of people without facial hair. The HyperBF networ ks perform satisfactorily on the classification tasks and exhibit the phenomenon of caricaturing, previously reported in psychophysical expe riments.