FG-Portrait: 3D Flow Guided Editable Portrait Animation
Yating Xu, Yunqi Miao, Evangelos Ververas, Jiankang Deng, Jifei Song · 2026
Read FG-Portrait: 3D Flow Guided Editable Portrait Animation by Yating Xu, Yunqi Miao, Evangelos Ververas, Jiankang Deng, Jifei Song (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-23381.txt.
Description
Xu and four co-authors animate portraits by moving motion from a driving video onto a still face. Diffusion tools often take only the driving motion, so source links get lost. Guessing dense 2D flow is unreliable. Their method computes 3D flow from parametric head models, with no learning for the flow itself. A 3D flow encoding lets each target pixel ask where it came from. Depth-guided sampling lines the flows up with real 2D motion. Users can also edit expressions and head pose by hand.
Catalog record
| CREATOR | Yating Xu, Yunqi Miao, Evangelos Ververas, Jiankang Deng, Jifei Song |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36342 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.23381