Orientation competition in cortical filters - an application to face recognition

A biologically motivated, computationally intensive approach to computer vision is developed and applied to the problem of automatic face recognition using two-dimensional Gabor functions which model the receptive eld functions of simple cells in the primary visual cortex of mammals.

Abstract

A biologically motivated, computationally intensive approach to computer vision is developed and applied to the problem of automatic face recognition. The approach is based on the use of twodimensional Gabor functions which model the receptive field functions of simple cells in the primary visual cortex of mammals. The convolutions of an input image with a set of antisymmetric visual receptive field functions (imaginary parts of Gabor functions) become the subject of thresholding and orientation competition. The developed cortical filters deliver highly structured information which is used for efficient feature extraction and representation in a lower dimension space. Applied to face recognition, the method gives a recognition rate of 98.5% on a large database of face images. 1 Introduction Large scale computer simulations are nowadays a well established research tool in natural and engineering sciences such as physics, chemistry, astronomy, fluid dynamics, electrical engineering, etc...