3D Face Recognition by Spatial Arrangement of iso-Geodesic Surfaces

An original framework for description and matching of three dimensional (3D) faces for recognition purposes is proposed in this paper. Face information is captured by extracting iso-geodesic surfaces of 3D face models. A compact representation of the faces is then constructed through a modeling technique capable to express the basic shape of iso-geodesic surfaces and quantitatively measure their spatial relationships in the 3D space. This information is encoded in an attributed relational graph. Experimental results on a 3D face database and baseline comparison show that the proposed solution attains high face recognition accuracy and is reasonably robust to facial expression changes. Experiments also target the identification of portions of the face surface which are most significant for the purpose of discriminating between faces.