An Illumination Insensitive Representation for Face Verification in the Frequency Domain

We present a comparison of several face representation methods from the point of view of their sensitivity to illumination changes. The sensitivity is measured in terms of the overlap of the distributions of normalized correlations used for inter class and intra class image comparisons. The result suggests, that better illumination invariance could be achieved in feature spaces in the frequency domain derived for a differentiated image rather than using the original input image.