3D Face Reconstruction from a Single 2D Image

Abstract

3D face reconstruction is the process of creating a 3D representation of a real human face. 3D face models have several applications like face recognition, 3D games, human-machine interaction, and plastic surgery simulations. Recently there has been a lot of research on deep learning methods for 3D face reconstruction from 2D face images. In this thesis, three deep learning-based methods for 3d reconstruction from a single image are reviewed. A new texturing method for 3D face models that uses the input photo as a UV texture image is proposed. Image warping is used to modify the input photo for this purpose. Warping is achieved using facial landmark detection and triangle meshes. A survey is conducted to assess the three face reconstruction methods and the proposed texturing method