DarkDiff: Advancing Low-Light Raw Enhancement by Retasking Diffusion Models for Camera ISP

TL;DR

DarkDiff leverages pre-trained diffusion models retasked with camera ISP to significantly improve the perceptual quality of low-light raw images, overcoming limitations of previous methods in detail and color accuracy.

Contribution

We propose a novel framework that retasks pre-trained diffusion models with camera ISP for enhanced low-light raw image enhancement, achieving superior perceptual quality.

Findings

Outperforms state-of-the-art in low-light raw image benchmarks

Produces sharper details and more accurate colors

Demonstrates effectiveness across multiple challenging datasets

Abstract

High-quality photography in extreme low-light conditions is challenging but impactful for digital cameras. With advanced computing hardware, traditional camera image signal processor (ISP) algorithms are gradually being replaced by efficient deep networks that enhance noisy raw images more intelligently. However, existing regression-based models often minimize pixel errors and result in oversmoothing of low-light photos or deep shadows. Recent work has attempted to address this limitation by training a diffusion model from scratch, yet those models still struggle to recover sharp image details and accurate colors. We introduce a novel framework to enhance low-light raw images by retasking pre-trained generative diffusion models with the camera ISP. Extensive experiments demonstrate that our method outperforms the state-of-the-art in perceptual quality across three challenging low-light raw image benchmarks.