Facial Asymmetry: A New Biometric

Abstract

Human facial asymmetry has long been a critical factor for evaluations of attractiveness and expressions in psychology and anthropology, although most studies are carried out qualitatively. In this work, we investigate in depth the effect of statistical facial asymmetry measurement as a biometric under expression variations. Our findings demonstrate that the asymmetry of specific facial regions captures individual differences that are robust to variation in facial expression. More importantly, our experimental results show that facial asymmetry provides discriminating power orthogonal to conventional face identification methods. The synergy of combining facial asymmetry with conventional methods is evaluated. Our work appears to be the first to show quantitatively the power of facial asymmetry as a biometric.