UniFace++: Revisiting a Unified Framework for Face Reenactment and Swapping via 3D Priors.
Saved in:
| Title: | UniFace++: Revisiting a Unified Framework for Face Reenactment and Swapping via 3D Priors. |
|---|---|
| Authors: | Xu, Chao1 (AUTHOR) 21832066@zju.edu.cn, Qian, Yijie1 (AUTHOR) 22332148@zju.edu.cn, Zhu, Shaoting1 (AUTHOR) zhust@zju.edu.cn, Sun, Baigui2 (AUTHOR) sunbaigui85@gmail.com, Zhao, Jian3,4 (AUTHOR) zhaoj90@chinatelecom.cn, Liu, Yong1 (AUTHOR) yongliu@iipc.zju.edu.cn, Li, Xuelong3,4 (AUTHOR) xuelong_li@chinatelecom.cn |
| Source: | International Journal of Computer Vision. Jul2025, Vol. 133 Issue 7, p4538-4554. 17p. |
| Subjects: | Generative adversarial networks, Simplicity, Supervision |
| Abstract: | Face reenactment and swapping share a similar pattern of identity and attribute manipulation. Our previous work UniFace has preliminarily explored establishing a unification between the two at the feature level, but it heavily relies on the accuracy of feature disentanglement, and GANs are also unstable during training. In this work, we delve into the intrinsic connections between the two from a more general training paradigm perspective, introducing a novel diffusion-based unified method UniFace++. Specifically, this work combines the advantages of each, i.e., stability of reconstruction training from reenactment, simplicity and effectiveness of the target-oriented processing from swapping, and redefining both as target-oriented reconstruction tasks. In this way, face reenactment avoids complex source feature deformation and face swapping mitigates the unstable seesaw-style optimization. The core of our approach is the rendered face obtained from reassembled 3D facial priors serving as the target pivot, which contains precise geometry and coarse identity textures. We further incorporate it with the proposed Texture-Geometry-aware Diffusion Model (TGDM) to perform texture transfer under the reconstruction supervision for high-fidelity face synthesis. Extensive quantitative and qualitative experiments demonstrate the superiority of our method for both tasks. [ABSTRACT FROM AUTHOR] |
| Copyright of International Journal of Computer Vision is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| FullText | Links: – Type: pdflink Text: Availability: 1 |
|---|---|
| Header | DbId: egs DbLabel: Engineering Source An: 185781640 AccessLevel: 6 PubType: Academic Journal PubTypeId: academicJournal PreciseRelevancyScore: 0 |
| IllustrationInfo | |
| Items | – Name: Title Label: Title Group: Ti Data: UniFace++: Revisiting a Unified Framework for Face Reenactment and Swapping via 3D Priors. – Name: Author Label: Authors Group: Au Data: <searchLink fieldCode="AR" term="%22Xu%2C+Chao%22">Xu, Chao</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> 21832066@zju.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Qian%2C+Yijie%22">Qian, Yijie</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> 22332148@zju.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Zhu%2C+Shaoting%22">Zhu, Shaoting</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> zhust@zju.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Sun%2C+Baigui%22">Sun, Baigui</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> sunbaigui85@gmail.com</i><br /><searchLink fieldCode="AR" term="%22Zhao%2C+Jian%22">Zhao, Jian</searchLink><relatesTo>3,4</relatesTo> (AUTHOR)<i> zhaoj90@chinatelecom.cn</i><br /><searchLink fieldCode="AR" term="%22Liu%2C+Yong%22">Liu, Yong</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> yongliu@iipc.zju.edu.cn</i><br /><searchLink fieldCode="AR" term="%22Li%2C+Xuelong%22">Li, Xuelong</searchLink><relatesTo>3,4</relatesTo> (AUTHOR)<i> xuelong_li@chinatelecom.cn</i> – Name: TitleSource Label: Source Group: Src Data: <searchLink fieldCode="JN" term="%22International+Journal+of+Computer+Vision%22">International Journal of Computer Vision</searchLink>. Jul2025, Vol. 133 Issue 7, p4538-4554. 17p. – Name: Subject Label: Subjects Group: Su Data: <searchLink fieldCode="DE" term="%22Generative+adversarial+networks%22">Generative adversarial networks</searchLink><br /><searchLink fieldCode="DE" term="%22Simplicity%22">Simplicity</searchLink><br /><searchLink fieldCode="DE" term="%22Supervision%22">Supervision</searchLink> – Name: Abstract Label: Abstract Group: Ab Data: Face reenactment and swapping share a similar pattern of identity and attribute manipulation. Our previous work UniFace has preliminarily explored establishing a unification between the two at the feature level, but it heavily relies on the accuracy of feature disentanglement, and GANs are also unstable during training. In this work, we delve into the intrinsic connections between the two from a more general training paradigm perspective, introducing a novel diffusion-based unified method UniFace++. Specifically, this work combines the advantages of each, i.e., stability of reconstruction training from reenactment, simplicity and effectiveness of the target-oriented processing from swapping, and redefining both as target-oriented reconstruction tasks. In this way, face reenactment avoids complex source feature deformation and face swapping mitigates the unstable seesaw-style optimization. The core of our approach is the rendered face obtained from reassembled 3D facial priors serving as the target pivot, which contains precise geometry and coarse identity textures. We further incorporate it with the proposed Texture-Geometry-aware Diffusion Model (TGDM) to perform texture transfer under the reconstruction supervision for high-fidelity face synthesis. Extensive quantitative and qualitative experiments demonstrate the superiority of our method for both tasks. [ABSTRACT FROM AUTHOR] – Name: AbstractSuppliedCopyright Label: Group: Ab Data: <i>Copyright of International Journal of Computer Vision is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.) |
| PLink | https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=185781640 |
| RecordInfo | BibRecord: BibEntity: Identifiers: – Type: doi Value: 10.1007/s11263-025-02395-6 Languages: – Code: eng Text: English PhysicalDescription: Pagination: PageCount: 17 StartPage: 4538 Subjects: – SubjectFull: Generative adversarial networks Type: general – SubjectFull: Simplicity Type: general – SubjectFull: Supervision Type: general Titles: – TitleFull: UniFace++: Revisiting a Unified Framework for Face Reenactment and Swapping via 3D Priors. Type: main BibRelationships: HasContributorRelationships: – PersonEntity: Name: NameFull: Xu, Chao – PersonEntity: Name: NameFull: Qian, Yijie – PersonEntity: Name: NameFull: Zhu, Shaoting – PersonEntity: Name: NameFull: Sun, Baigui – PersonEntity: Name: NameFull: Zhao, Jian – PersonEntity: Name: NameFull: Liu, Yong – PersonEntity: Name: NameFull: Li, Xuelong IsPartOfRelationships: – BibEntity: Dates: – D: 01 M: 07 Text: Jul2025 Type: published Y: 2025 Identifiers: – Type: issn-print Value: 09205691 Numbering: – Type: volume Value: 133 – Type: issue Value: 7 Titles: – TitleFull: International Journal of Computer Vision Type: main |
| ResultId | 1 |
