A model of face space

Abstract

Faces vary in their distinctiveness, and it has been suggested that—more distinctive faces are located further from the mean face in internal or perceptual face space (Valentine, 1991), than typical faces. This paper proposes a pixel-based computational model of face distinctiveness. I show that the notion of psychological distinctiveness is correlated with the metrical Euclidean distance, from the mean face. Distances are measured in pixel-based face space, which serves as the basis of the computational model. A prediction of the model is that pairwise perceptual distances between distinctive faces are, on average, greater than between typical faces. This is affirmed by psychophysical experiment. This finding confirms a key axiom of the norm-based coding model, which states that face space is centrally distributed with typical faces packed densely near the centre of the space and distinctive faces spread sparsely in the periphery. This assumption has previously been confirmed by Davidenko (2007) using face silhouettes, and here I show that this axiom also generalizes to full frontal face images.