Wavelet-Guided Semantic Signal Compensation for Inversion-Free Image Editing
2026-07-02 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors study a way to edit images based on text prompts without reversing the image creation process (inversion). They found that in some cases, their method (FlowEdit) doesn't change the image enough early on, especially for big overall changes. They suggest this happens because the model focuses too much on typical image patterns, ignoring the text direction. To fix this, they introduce a new technique that emphasizes important editing signals early while keeping the background mostly the same, improving global changes without ruining the image’s structure.
text-guided image editinginversion-free editingFlowEditsemantic changesimage generationdata manifoldfrequency-aware compensationbackground preservationnoise regimeediting trajectory
Authors
Anqi Tang, Wenhao Sun, Zhaoqiang Liu
Abstract
Text-guided image editing aims to modify visual content according to a target prompt while preserving the background. Recent inversion-free image editing frameworks such as FlowEdit have demonstrated strong editing capability without requiring inversion. Empirically, FlowEdit can achieve substantial semantic changes under appropriate hyperparameter settings. However, we observe that under certain global attribute shifts, the editing trajectory may not effectively move away from the source distribution in the early timesteps. Our analysis suggests that in the high-noise regime, the dominant manifold-seeking flow toward the data manifold can reduce the influence of the text-conditioned direction, leading to limited global modification while background structures remain only moderately preserved. Inspired by this observation, we propose an inversion-free, frequency-aware semantic compensation strategy that strengthens the effective signal in the early stage of generation, while maintaining structural consistency in the background. The proposed method improves global editing capacity without sacrificing background fidelity.