ReFlex: Text-Guided Editing of Real Images in Rectified Flow via Mid-Step Feature Extraction and Attention Adaptation

TL;DR

This paper introduces ReFlex, a training-free, text-guided real-image editing method that leverages intermediate features and attention adaptation in ReFlow, achieving superior results without masks or source prompts.

Contribution

ReFlex is the first method to enable training-free, mask-free, and prompt-free real-image editing in ReFlow by analyzing mid-step features and adapting attention, improving alignment and editability.

Findings

Outperforms nine baselines on two benchmarks

Human evaluations favor ReFlex's editing quality

Effective without user masks or source prompts

Abstract

Rectified Flow text-to-image models surpass diffusion models in image quality and text alignment, but adapting ReFlow for real-image editing remains challenging. We propose a new real-image editing method for ReFlow by analyzing the intermediate representations of multimodal transformer blocks and identifying three key features. To extract these features from real images with sufficient structural preservation, we leverage mid-step latent, which is inverted only up to the mid-step. We then adapt attention during injection to improve editability and enhance alignment to the target text. Our method is training-free, requires no user-provided mask, and can be applied even without a source prompt. Extensive experiments on two benchmarks with nine baselines demonstrate its superior performance over prior methods, further validated by human evaluations confirming a strong user preference for our approach.