Face^3 a 2D+3D Robust Face Recognition System

Abstract

We present Face3, a system for the automatic recognition of people using 2D and 3D images. The system implements several innovative algorithms. A full 3D face detector allows the detection effaces regardless their pose and orientation in the scene. A face restoration module allows the recognition even in presence of occlusions. The recognition is performed using a fusion of 2D and 3D data. Finally, in order to relax the hardware constraints, a simplification algorithm is used to reduce the data complexity of the 3D face models.