Video-based face recognition

Abstract

Video-based face recognition has been one of the hot topics in the field of pattern recognition in the last several decades. In this paper, we put forward a novel method, which is named as V-LPP. Our method uses Locality Preserving Projections (LPP) to recognize video-based face sequence, so it can discover more space-time semantic information hidden in video face sequence, simultaneously make full use of the intrinsic nonlinear structure information to extract discriminative manifold features. In the end, we also compare our V-LPP with Eigenfaces (PCA), Fisherfaces (LDA) and Laplacianfaces (LPP) on UCSD/Honda Video Database. Experimental results suggest that V-LPP can provide a more satisfying representation and achieve lower error rates in video-based face recognition.