A novel semi-supervised face recognition for video

Abstract

Video-based face recognition has been one of the hot topics in the field of pattern recognition in the last few decades. In this paper, incorporating Support Vector Machines (SVM) and Locality Preserving Projections (LPP), we propose a novel semi-supervised face recognition algorithm for video, which can discover more space-time semantic information hidden in video face sequence, simultaneously make full use of the small amount of labeled data with the plentiful unknown information and the intrinsic nonlinear structure information to extract discriminative manifold features. We also compare our algorithm with other algorithms on UCSD/Honda Video Database. The experimental results show that the proposed algorithm can outperform state-of-the-art solutions for videobased face recognition.