Region-Aware Face Swapping

Chao Xu, Jiangning Zhang, Miao Hua, Qian He, Zili Yi, Yong Liu

CVPR 2022 pp. 7632-7641

doi:10.1109/CVPR52688.2022.00748 /cvpr/2022/xu2022cvpr-regionaware/

Abstract

This paper presents a novel Region-Aware Face Swapping (RAFSwap) network to achieve identity-consistent harmonious high-resolution face generation in a local-global manner: 1) Local Facial Region-Aware (FRA) branch augments local identity-relevant features by introducing the Transformer to effectively model misaligned cross-scale semantic interaction. 2) Global Source Feature-Adaptive (SFA) branch further complements global identity-relevant cues for generating identity-consistent swapped faces. Besides, we propose a Face Mask Predictor (FMP) module incorporated with StyleGAN2 to predict identity-relevant soft facial masks in an unsupervised manner that is more practical for generating harmonious high-resolution faces. Abundant experiments qualitatively and quantitatively demonstrate the superiority of our method for generating more identity-consistent high-resolution swapped faces over SOTA methods, e.g., obtaining 96.70 ID retrieval that outperforms SOTA MegaFS by 5.87.

PDF CVPR Semantic Scholar

Cite

Text

Xu et al. "Region-Aware Face Swapping." Conference on Computer Vision and Pattern Recognition, 2022. doi:10.1109/CVPR52688.2022.00748

Markdown

[Xu et al. "Region-Aware Face Swapping." Conference on Computer Vision and Pattern Recognition, 2022.](https://mlanthology.org/cvpr/2022/xu2022cvpr-regionaware/) doi:10.1109/CVPR52688.2022.00748

BibTeX

@inproceedings{xu2022cvpr-regionaware,
  title     = {{Region-Aware Face Swapping}},
  author    = {Xu, Chao and Zhang, Jiangning and Hua, Miao and He, Qian and Yi, Zili and Liu, Yong},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2022},
  pages     = {7632-7641},
  doi       = {10.1109/CVPR52688.2022.00748},
  url       = {https://mlanthology.org/cvpr/2022/xu2022cvpr-regionaware/}
}