Face Recognition based on Accelerated Joint Boosting and Illumination Normalized Local Gabor Binary Pattern Histogram Sequence

Determining what features are optimal for face representation is quite a challenge in Face Recognition. Joint Boosting is a strong algorithm to select important features from feature pool. And, it outperforms other feature selection methods. But it takes quite a long time to train on a large training set. In this paper, an Accelerated Joint Boosting is proposed to resolve the time wasting problem. The contributions of this paper are: (1) We propose an Accelerated Joint Boosting method for feature selection in Face Recognition. (2) An efficient preprocessing chain for Illumination Normalization which is called 'PP' and Histogram Sequence of Local Gabor Binary Pattern (LGBPHS) are combined for robust Face Recognition, and it gets remarkable performance in Face Recognition Grand Challenge version 2.0 experiment 4 (FRGC).