Metrological performance comparison of biometric system architectures for 3D face recognition

Different biometric system architectures based on 3D images can be realized for face recognition. In such systems the triangulation of images provided by a couple of 2D cameras is employed to achieve the 3D features of a face. To setting up the system, different positioning of cameras as well as both kind and resolution of camera could be considered. These parameters can affect the correct decision rate of the system in classifying the input face, especially in presence of image uncertainty. In previous papers, the authors proposed an original approach for the estimation of the confidence level of results provided by classification systems for face recognition. Such approach is here adopted in order to compare several 3D architectures differing in camera specifications and geometrical positioning.