Maximum variance projections for face recognition

Authors:
  • DZDeli Zhao
    Deli Zhao

Abstract
Maximum variance projection (MVP), as a novel subspace learning algorithm, is proposed. It is a linear discriminant algorithm that preserves local information by capturing the local geometry of the manifold. Two abilities of manifold learning and classification are combined into the properties of our algorithm. Since face images often belong to a submanifold of intrinsically low dimension, we carry out the MVP algorithm for face manifold learning and classification. Several experiments show the effectiveness of our developed algorithm.

This publication has 18 references indexed in Scilit: