A facial aging approach to identification of identical twins

Abstract

Biometric recognition based on the characteristics of human faces has attracted a great deal of attention over the past few years. However, the similarity in the facial appearance of identical twins has made the task difficult and has even compromised commercial face recognition systems. In this paper, we shed new light on the study of facial recognition of identical twins and propose a novel approach using twins group classification and facial aging features to tell them apart. Our experiments, conducted on the University of Notre Dame ND-twins database, that was acquired over two years (2009 and 2010), indicate that our proposed approach demonstrates good generalization ability and high identification rates.