Features for robust face-based identity verification

In this paper we propose the discrete cosine transform (DCT) mod 2 feature set, which utilizes polynomial coe2cients derived from 2D DCT coe2cients obtained from spatially neighboring blocks. Face veri%cation results on the multi-session VidTIMIT database suggest that the DCT-mod 2 feature set is superior (in terms of robustness to illumination direction changes and discrimination ability) to features extracted using three popular methods: eigenfaces principal component analysis, 2D DCT and 2D Gabor wavelets. Moreover, compared to Gabor wavelets, the DCT-mod 2 feature set is over 80 times faster to compute. Additional experiments on the Weizmann database also show that the DCT-mod 2 approach is more robust than 2D Gabor wavelets and 2D DCT coe2cients. ? 2003 Elsevier Science B.V. All rights reserved.