3D face recognition based on matching of facial surfaces
B. A. Echeagaray-Patrón, Vitaly Kober · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 2015
Face recognition is an important task in pattern recognition and computer vision. In this work a method for 3D face recognition in the presence of facial expression and poses variations is proposed. The method uses 3D shape data without color or texture information. A new matching algorithm based on conformal mapping of original facial surfaces onto a Riemannian manifold followed by comparison of conformal and isometric invariants computed in the manifold is suggested. Experimental results are presented using common 3D face databases that contain significant amount of expression and pose variations.