An Efficiently 3D Face Recognizing Method Using Range Image and Sparse Representation

Abstract

In order to eliminate the effect of the facial expression and illumination condition, as well as to speed up the recognition procedure, we propose a face recognition approach based on sparse representation. First, preprocessing and segmenting the face area from three dimensional (3D) face scans, we also apply coarse to fine registration to ensure the alignment of range images; second, mapping the 3D model to range image through a kind of geometry-based resampling method; finally, employ sparse representation classification method to identify 3D face. The experiment results in actual 3D face database demonstrate the effectiveness of the proposed method.