UniEdit-Flow: Unleashing Inversion and Editing in the Era of Flow Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jiao, Guanlong, Huang, Biqing, Wang, Kuan-Chieh, Liao, Renjie
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910237506666496
author Jiao, Guanlong
Huang, Biqing
Wang, Kuan-Chieh
Liao, Renjie
author_facet Jiao, Guanlong
Huang, Biqing
Wang, Kuan-Chieh
Liao, Renjie
contents Flow matching models have emerged as a strong alternative to diffusion models, but existing inversion and editing methods designed for diffusion are often ineffective or inapplicable to them. The straight-line, non-crossing trajectories of flow models pose challenges for diffusion-based approaches but also open avenues for novel solutions. In this paper, we introduce a predictor-corrector-based framework for inversion and editing in flow models. First, we propose Uni-Inv, an effective inversion method designed for accurate reconstruction. Building on this, we extend the concept of delayed injection to flow models and introduce Uni-Edit, a region-aware, robust image editing approach. Our methodology is tuning-free, model-agnostic, efficient, and effective, enabling diverse edits while ensuring strong preservation of edit-irrelevant regions. Extensive experiments across various generative models demonstrate the superiority and generalizability of Uni-Inv and Uni-Edit, even under low-cost settings. Project page: https://uniedit-flow.github.io/
format Preprint
id arxiv_https___arxiv_org_abs_2504_13109
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle UniEdit-Flow: Unleashing Inversion and Editing in the Era of Flow Models
Jiao, Guanlong
Huang, Biqing
Wang, Kuan-Chieh
Liao, Renjie
Computer Vision and Pattern Recognition
Flow matching models have emerged as a strong alternative to diffusion models, but existing inversion and editing methods designed for diffusion are often ineffective or inapplicable to them. The straight-line, non-crossing trajectories of flow models pose challenges for diffusion-based approaches but also open avenues for novel solutions. In this paper, we introduce a predictor-corrector-based framework for inversion and editing in flow models. First, we propose Uni-Inv, an effective inversion method designed for accurate reconstruction. Building on this, we extend the concept of delayed injection to flow models and introduce Uni-Edit, a region-aware, robust image editing approach. Our methodology is tuning-free, model-agnostic, efficient, and effective, enabling diverse edits while ensuring strong preservation of edit-irrelevant regions. Extensive experiments across various generative models demonstrate the superiority and generalizability of Uni-Inv and Uni-Edit, even under low-cost settings. Project page: https://uniedit-flow.github.io/
title UniEdit-Flow: Unleashing Inversion and Editing in the Era of Flow Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2504.13109