Authors: Yujuan Sun; Xiaofeng Zhang; Muwei Jian
Addresses: School of Information and Electrical Engineering, Ludong University, Yantai, China ' School of Information and Electrical Engineering, Ludong University, Yantai, China ' Department of Computer Science and Technology, Ocean University of China, Qingdao, China
Abstract: The three-dimensional reconstruction from single input image is quite difficult due to many unknown parameters, such as the light conditions, the surface normal and albedo of the object. However, there are overall similar characteristics for different human faces, such as the shapes and the positions of the eyes, nose; mouth and ears are generally identical. The similar characteristics has been used in this paper to relax the numbers of the input face images, and reconstruct the 3D shape based on a couple statistical model. Moreover, the light condition of the single input image can be different from that of training database. The experiment results show the effectiveness of the proposed method.
Keywords: three-dimensional reconstruction; coupled statistical model; human face.
International Journal of Information and Computer Security, 2018 Vol.10 No.2/3, pp.248 - 260
Received: 12 Dec 2016
Accepted: 17 Feb 2017
Published online: 23 Apr 2018 *