Improving Diffusion-Based Image Editing Faithfulness via Guidance and Scheduling

TL;DR

This paper introduces Faithfulness Guidance and Scheduling (FGS), a method to improve the faithfulness of diffusion-based image editing while preserving editability, enabling more precise and high-quality image modifications.

Contribution

The paper proposes FGS, a novel approach that enhances faithfulness in diffusion-based image editing with minimal impact on editability, addressing the trade-off between these aspects.

Findings

FGS improves faithfulness in image editing.

FGS maintains high editability alongside faithfulness.

Compatible with various editing methods for diverse tasks.

Abstract

Text-guided diffusion models have become essential for high-quality image synthesis, enabling dynamic image editing. In image editing, two crucial aspects are editability, which determines the extent of modification, and faithfulness, which reflects how well unaltered elements are preserved. However, achieving optimal results is challenging because of the inherent trade-off between editability and faithfulness. To address this, we propose Faithfulness Guidance and Scheduling (FGS), which enhances faithfulness with minimal impact on editability. FGS incorporates faithfulness guidance to strengthen the preservation of input image information and introduces a scheduling strategy to resolve misalignment between editability and faithfulness. Experimental results demonstrate that FGS achieves superior faithfulness while maintaining editability. Moreover, its compatibility with various editing methods enables precise, high-quality image edits across diverse tasks.