A unified 3D face authentication framework based on robust localmesh SIFT feature

In this paper, we design a unified 3D face authentication system for practical use. First, we propose afacial depth recovery method to construct a facial depth map from stereoscopic videos. It effectivelyutilize prior facial information and incorporate the visibility term to classify static and dynamic pixels forrobust depth estimation. Secondly, in order to make 3D face authentication more accurate and consistent,we present an intrinsic scale feature detection for interesting points on 3D facial mesh regions.Then, a novel feature descriptor is proposed, called Local Mesh Scale-Invariant Feature Transform(LMSIFT) to reflect the different face recognition abilities in different facial regions. Finally, the sparseoptimization problem of visual codebook is used to 3D face learning. We evaluate our approach onpublicly available 3D face databases and self-collected realistic scene databases. We also develop aninteractive education system to investigate its performance in practice, which demonstrates the highperformance of the proposed approach for accurate 3D face authentication. Compared with previouspopular approaches, our system has consistently better performance in terms of effectiveness, robustnessand universality.