Two-dimensional Intra-class Diversity Preserving for Face Recognition
Wang Cheng-ru · Guangdian gongcheng · 2012
In order to make Enhanced Fisher Disriminant Criterion(EFDC) avoid of impairing the discriminating information caused by PCA dimension reduction,based on EFDC,a method of two-dimensional Intra-class Diversity Preserving(2D-IDP) for face recognition is proposed.In this method,we built a more robust discriminate criterion which can make the data points of different class as distant as possible and simultaneously preserve the intra-class compactness and variation,and thus avoid the over-fitting problem.At the same time,we redefined parameter t in the neighbor graph of EFDC which made it change adaptively according to different samples.Thus it avoided the problem of performance degradation caused by inappropriate choice of t.Experiments on YALE and AR face database verify the effectiveness of the proposed method.