SVM-based Discriminant Analysis for face recognition
Sang‐Ha Kim, Kar‐Ann Toh, Sangyoun Lee · 2008
In this paper, we introduce a novel variant of Linear Discriminant Analysis (LDA) for face recognition. The proposed method attempts to find an optimal LDA matrix by redesigning the between-class scatter matrix incorporating a Support Vector Machine (SVM). Our empirical evaluations show that the proposed method offers noticeable performance improvement over the conventional LDA.