SpotEdit: Selective Region Editing in Diffusion Transformers

TL;DR

SpotEdit introduces a training-free diffusion editing framework that selectively updates only modified regions, reducing computation and preserving unaltered areas for efficient and high-quality image editing.

Contribution

It proposes a novel selective editing method using SpotSelector and SpotFusion, eliminating the need for retraining and improving editing efficiency and fidelity.

Findings

Reduces unnecessary computation during image editing.

Maintains high fidelity in unmodified regions.

Achieves efficient and precise editing without retraining.

Abstract

Diffusion Transformer models have significantly advanced image editing by encoding conditional images and integrating them into transformer layers. However, most edits involve modifying only small regions, while current methods uniformly process and denoise all tokens at every timestep, causing redundant computation and potentially degrading unchanged areas. This raises a fundamental question: Is it truly necessary to regenerate every region during editing? To address this, we propose SpotEdit, a training-free diffusion editing framework that selectively updates only the modified regions. SpotEdit comprises two key components: SpotSelector identifies stable regions via perceptual similarity and skips their computation by reusing conditional image features; SpotFusion adaptively blends these features with edited tokens through a dynamic fusion mechanism, preserving contextual coherence and editing quality. By reducing unnecessary computation and maintaining high fidelity in unmodified areas, SpotEdit achieves efficient and precise image editing.