3D depth analysis of human faces

Abstract

We provide an important analysis of depth variation of human faces. Throughout an extensive analysis of 3D face shapes, we claim that 3D depth information (z) of faces is not significantly changing and can be synthesized from another person's depth or a generic depth information. We also show that gender and ethnicity specific average depth models can approximate the 3D shape of the input face image more accurately, achieving a better generalization of 3D face modeling and reconstruction compared to a global average depth model.