DirectEdit: Step-Level Accurate Inversion for Flow-Based Image Editing

TL;DR

DirectEdit offers a novel flow-based image editing method that achieves precise inversion and high-fidelity editing without additional neural evaluations, outperforming existing techniques.

Contribution

It introduces a direct alignment approach for inversion in flow models, eliminating reconstruction errors and improving image editing accuracy and efficiency.

Findings

Outperforms state-of-the-art image editing methods in accuracy.

Achieves precise inversion without extra neural function evaluations.

Balances fidelity and editability effectively.

Abstract

With recent advancements in large-scale pre-trained text-to-image (T2I) models, training-free image editing methods have demonstrated remarkable success. Typically, these methods involve adding noise to a clean image via an inversion process, followed by separate denoising steps for the reconstruction and editing paths during the forward process. However, since the reconstruction path is approximated using noisy latents from mismatched timesteps, existing methods inevitably suffer from accumulated drift, which fundamentally limits reconstruction fidelity. To address this challenge, we systematically analyze the inversion process within the flow transformer and propose DirectEdit, a simple yet effective editing method that eliminates the inherent reconstruction error without introducing additional neural function evaluations (NFEs). Unlike most prior works that attempt to rectify the inversion path, DirectEdit focuses on directly aligning the forward paths, enabling precise reconstruction and reliable feature sharing. Furthermore, we introduce a preservation mechanism based on attention feature injection and multi-branch mask-guided noise blending, which effectively balances fidelity and editability. Extensive experiments across diverse scenarios demonstrate that DirectEdit achieves efficient and accurate image editing, delivering superior performance that outperforms state-of-the-art methods. Code and examples are available at https://desongyang.github.io/Directedit.