SL2E-AFRE : Personalized 3D face reconstruction using autoencoder with simultaneous subspace learning and landmark estim
- PDF / 7,711,513 Bytes
- 16 Pages / 595.224 x 790.955 pts Page_size
- 92 Downloads / 163 Views
SL2E-AFRE : Personalized 3D face reconstruction using autoencoder with simultaneous subspace learning and landmark estimation P. R. Suganya Devi1
· R. Baskaran1
Accepted: 1 October 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract 3D face reconstruction from single face image has received much attention in the past decade, as it has been used widely in many applications in the field of computer vision. Despite more accurate solutions by 3D scanners and several commercial systems, they have drawbacks such as the need for manual initialization, time and economy constraints. In this paper, a novel framework for 3D face reconstruction is presented. Firstly, landmarks are localized on the database faces with the proposed landmark-mapping strategy employing a model template. Then, an autoencoder assisted by the proposed energy function to simultaneously learn the facial patch subspace and the keypoints positions is employed to predict the landmarks. Finally, an unique 3D reconstruction is obtained with the proposed predicted landmark based deformation. Meta-parameters are incorporated into the energy function during the training phase to enhance the performance of the autoencoder network in reconstructing the face model. The experiments are carried out on two databases namely the USF Human ID 3-D Database and the Bosphorus 3D face database. The experimental results show that the Autoencoder based Face REconstruction with Simultaneous patch Learning and Landmark Estimation method (SL2E-AFRE) is efficient and the performance of the same is significantly upgraded in each iteration. Keywords 3D Face reconstruction · Autoencoder · Landmark estimation · Shape deformation
1 Introduction Three-dimensional knowledge about a human face is very much beneficial for computer vision and computer animation fields. 3D Face Reconstruction can be defined as promoting a 2D face image into a 3D geometry. Since a 3D face model provides a mathematical representation of the face surface in three dimension, which is consistent despite change in the pose, illumination and expression, 3D face reconstruction is therefore essential for real-time applications in face recognition, plastic surgery, facial animation [10, 29], entertainment [9] and computer games [38, 39], 3D rendering of comic cartoon characters [26, 27] etc. In the media and entertainment industry(M&E), P. R. Suganya Devi
[email protected] R. Baskaran [email protected] 1
Department of Computer Science and Engineering, College of Engineering, Guindy, Anna University, Chennai, India
it is contemporary to use 3D rendering of human faces in various segments such as cinema, advertising and gaming. In the medical field, 3D printing exhibits abundant applications such as 3D modeling of human organs from computed tomography(CT) data [43]. 3D reconstruction is essential to generate 3D-printed model of the anatomical structure for pre-surgical preparation and also to carry out cosmetic surgery. In criminal investigation [14, 32, 41], facial reco
Data Loading...
