New 3D face matching technique for 3D model based face recognition

Abstract

Various methods have been used for face recognition over the past few years. The motivation for the continuous work on face recognition is to obtain a method which is able to recognize different angles and poses of faces accurately and efficiently. Currently, faces are identified using either two dimensional (2D) images or three dimensional (3D) range images. In this paper, a face recognition method that is able to recognize faces at various angles is proposed. This method uses only the three dimensional range images for matching. Firstly, surface matching, which consists of calculating the surface distance of the probe face with the faces in the database, is performed on the aligned face curves. The top ten candidates from the surface matching are then further processed using Principal Component Analysis (PCA) followed by Linear Discriminant Analysis (LDA). The database candidate with the lowest Euclidean distance value will be identified as the probe face. When compared with the multiview method of face recognition, which uses two dimensional images, the proposed method is able to obtain higher recognition rates. The method proposed is a fully automatic face recognition system.