Biometric facial verification using 2D and 3D information

The objective of this paper is to review the work in a recent thesis devoted to the design, development and test of a multimodal face verification system than combines bidimensional and threedimensional information. A deeply study of the discrimination capacity of 2D or 3D information was made. It has been designed and developed three different face verification systems, based on different data: texture image (2D), range data (2.5D) and three dimensional mesh (3D). The 2D and 2.5D verifiers are based on the principal component analysis and support vector machine. The 3D verifier it is based on the measure of the quality of surfaces adjustment using the algorithm iterative closest point. An automatic 3D facial feature location method was designed too based on the spin images technique.