RSFace: subject agnostic face swapping with expression high fidelity
Article
Article Title | RSFace: subject agnostic face swapping with expression high fidelity |
---|---|
ERA Journal ID | 18149 |
Article Category | Article |
Authors | Yang, Gaoming, Wang, Tao, Fang, Xianjin and Zhang, Ji |
Journal Title | The Visual Computer |
Journal Citation | 39 (11), pp. 5497-5511 |
Number of Pages | 15 |
Year | 2023 |
Publisher | Springer |
Place of Publication | Germany |
ISSN | 0178-2789 |
1432-2315 | |
Digital Object Identifier (DOI) | https://doi.org/10.1007/s00371-022-02675-z |
Web Address (URL) | https://link.springer.com/article/10.1007/s00371-022-02675-z |
Abstract | Face swapping has shown remarkable progress with the flourishing development of deep learning. In particular, the emergence of subject agnostic methods has broadened the range of applications of face swapping. Furthermore, high fidelity implementation has improved the naturalness of generated faces. However, some high fidelity face swapping methods still suffer from expression distortion at this stage. In this work, we propose an extended Adaptive Embedding Integration Network (AEI-Net) to improve the performance of this network in synthesizing swapped faces on faces in the wild. First, we add a face reenactment module to synchronize the expressions of the input faces and reduce the influence of irrelevant attributes on the synthesis results. Second, we train AEI-Net using a new attribute matching loss to improve the consistency of the generated results and the target face expressions. Finally, extensive experiments on wild faces demonstrate that our method can better restore expression and posture while maintaining identity than previous methods. |
Keywords | Attribute matching loss ; Face swapping ; Face reenactment ; High fidelity |
ANZSRC Field of Research 2020 | 460299. Artificial intelligence not elsewhere classified |
Public Notes | Files associated with this item cannot be displayed due to copyright restrictions. |
Byline Affiliations | Anhui University of Science and Technology, China |
School of Mathematics, Physics and Computing |
https://research.usq.edu.au/item/z02z0/rsface-subject-agnostic-face-swapping-with-expression-high-fidelity
48
total views0
total downloads0
views this month0
downloads this month