InstantRetouch: Personalized Image Retouching without Test-time Fine-tuning Using an Asymmetric Auto-Encoder
Temesgen Muruts Weldengus, Binnan Liu, Fei Kou, Youwei Lyu, Jinwei Chen, Qingnan Fan, Changqing Zou, [4pt] Zhejiang University, [email protected] · 2026
Read InstantRetouch: Personalized Image Retouching without Test-time Fine-tuning Using an Asymmetric Auto-Encoder by Temesgen Muruts Weldengus, Binnan Liu, Fei Kou, Youwei Lyu, Jinwei Chen, Qingnan Fan, Changqing Zou, [4pt] Zhejiang University, [email protected] (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-2602-17044.txt.
Description
Temesgen Muruts Weldengus led a seven-author team on a tool that retouches photos in a user's own style, with no per-user fine-tuning at test time. Current personal retouching tools often demand that tuning step, or they fail to carry a style to new images well. The paper introduces RefRetouch, a general framework that reads a style from paired reference examples and applies it at once.
An asymmetric auto-encoder turns the reference pairs into a style code stripped of image content, so the style can move cleanly to a new photo. A retrieval step, called retrieval-augmented retouching, then pulls style codes from the references closest in content to the photo at hand. Tests cover single-reference, multi-reference, and mixed-style cases, plus photorealistic style transfer, where the tool also held up.
Catalog record
| CREATOR | Temesgen Muruts Weldengus, Binnan Liu, Fei Kou, Youwei Lyu, Jinwei Chen, Qingnan Fan, Changqing Zou, [4pt] Zhejiang University, [email protected] |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 47 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-17044