Expression Invariant Face Recognition Using Biologically Inspired Features from Visual Cortex Mechanism

摘要

Face recognition performance has always been influenced by the various facial expressions a subject may achieve. In this paper, we investigate face as a biometric under expression variation using Biologically Inspired Features (BIFs). We apply biologically inspired features, derived from a feed forward model of object recognition pathway in visual cortex for expression invariant face recognition problem. In this work, our goal is to understand whether features motivated by a model of visual cortex are robust for human identification under expression variation. Experimental results show that the model achieves high recognition percentage even for large expression variations.