Face-tree: A compact discrete 3D face shape representation

Recently, there has been an increasing interest in using 3D facial images as biometric modality. To that end, a proper representation of the 3D facial shape that would allow effective face matching is a major requirement. In this paper, we propose a novel 3D facial representation, dubbed, the face-tree. We describe the extraction process of this representation from 3D triangular mesh face model, then we showcase its distinguished properties which include, ordered structure, compactness, richness, efficiency, and capacity for partial deployment. We discuss the potential of the face-tree representation for 3D face recognition, through experiments made with two different database sets.