Multi-view self-supervised learning for 3D facial texture reconstruction from single image

作者：

Highlights：

• A multi-view self-supervised deep network for 3D facial texture reconstruction from a single image is proposed.

• A multi-view consistency loss function was proposed to train the self-supervised network.

• Our method performs favorably against the recent representative methods.

摘要

•A multi-view self-supervised deep network for 3D facial texture reconstruction from a single image is proposed.•A multi-view consistency loss function was proposed to train the self-supervised network.•Our method performs favorably against the recent representative methods.

论文关键词：Deep learning,3D face reconstruction,UV texture reconstruction,Convolutional neural networks

论文评审过程：Received 4 September 2020, Revised 29 August 2021, Accepted 11 September 2021, Available online 20 September 2021, Version of Record 27 September 2021.

论文官网地址：https://doi.org/10.1016/j.imavis.2021.104311