FCAT-Diff: Flexible and Consistent Appearance Transfer Based on Training-free Diffusion Model

Open Webpage

Zhengyi Gong, Mingwen Shao, Chang Liu, Xiang Lv, Huan Liu

Published: 01 Aug 2025, Last Modified: 16 Nov 2025Computers & GraphicsEveryoneRevisionsCC BY-SA 4.0
Abstract: Highlights•A training-free framework for flexible, consistent appearance transfer.•SOTA performance in preserving structural and background consistency.•Allowing region-specific and multi-appearance transfer.