High-Quality Exposure Correction with Diffusion-Based Image Generation Priors

2026-08-09

AI Summary

Existing exposure correction methods often overemphasize pixel-level fidelity, struggling with extreme exposure regions and consequently degrading perceptual quality. This work proposes DPEC, a novel framework that, for the first time, effectively leverages the generative priors of large-scale pre-trained diffusion models for this task. By employing efficient fine-tuning, DPEC achieves single-step denoising while integrating the strengths of regression-based approaches. Its core innovations include a single-step denoising mechanism and a joint cross-attention module that fuses multi-scale diffusion prior features, preserving high-frequency details without introducing stochastic artifacts. Extensive experiments demonstrate that DPEC consistently outperforms state-of-the-art methods across multiple benchmark datasets in terms of fidelity, perceptual quality, and visual appeal.

Abstract

Although most existing exposure correction methods achieve high fidelity, they often place excessive focus on overall pixel-wise accuracy, making it challenging to effectively model extreme exposure regions, which results in suboptimal perceptual quality. Recently, diffusion models have received significant attention due to their remarkable performance in the realm of image generation. However, their successful application to exposure correction remains a challenging and open question. The key challenge lies in generating accurate image structures and maintaining high image fidelity during stochastic diffusion processes. In this paper, we propose DPEC (Diffusion Prior-based Exposure Correction), a novel framework for image exposure correction that utilizes diffusion-based image generation priors encapsulated in pre-trained large-scale diffusion models. Specifically, we first propose an efficient fine-tuning strategy to derive an exposure corrector from pre-trained models, enabling the generation of enhanced images in a single-step denoising process. Moreover, we seamlessly combine the strengths of diffusion models and regression models, and design a joint cross-attention module to integrate multi-scale diffusion prior features, thereby effectively preserving high-frequency details and minimizing random artifacts. The diffusion model focuses on dealing with low-frequency content rather than all the intricate texture details. The experimental results demonstrate that the proposed DPEC method consistently outperforms existing state-of-the-art methods on multiple exposure correction datasets, whether in terms of fidelity, perceptual quality, or visual effects.

Problem

• exposure correction
• diffusion models
• perceptual quality
• image fidelity
• extreme exposure regions

Innovation

• diffusion prior
• exposure correction
• single-step denoising
• cross-attention fusion
• perceptual quality