FramePainter: Endowing Interactive Image Editing with Video Diffusion Priors
Yabo Zhang Xinpeng Zhou Yihan Zeng Hang Xu Hui Li Wangmeng Zuo · 2025
Read FramePainter: Endowing Interactive Image Editing with Video Diffusion Priors by Yabo Zhang Xinpeng Zhou Yihan Zeng Hang Xu Hui Li Wangmeng Zuo (2025) 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-2501-08225.txt.
Description
Yabo Zhang and co-authors present a tool for editing images by dragging and drawing, in a paper from 2025. Such editing lets a user pull a point in a photo and have the scene change in a lifelike way. Past tools learned this from video clips, but they were built on still image models and needed huge training sets.
The new tool, FramePainter, treats editing as a short video task instead. It starts from a video generation model, so real world motion patterns come built in. A light control part feeds the user's edit signals into the model. That design cuts training cost and helps the edit stay steady across frames. The authors test the tool on standard editing tasks against past methods.
Catalog record
| CREATOR | Yabo Zhang Xinpeng Zhou Yihan Zeng Hang Xu Hui Li Wangmeng Zuo |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-08225