Personalized Generative Low-light Image Denoising and Enhancement

TL;DR

This paper introduces DiffPGD, a personalized diffusion model for low-light image denoising that leverages user photo galleries and an identity-consistent physical buffer to improve image restoration without fine-tuning.

Contribution

The paper presents a novel personalized diffusion-based denoising method that incorporates an identity-consistent physical buffer for improved low-light image enhancement.

Findings

Outperforms existing diffusion-based denoising methods in low-light scenarios

Uses personalized photo galleries to tailor the model to individual users

Achieves superior denoising and enhancement without fine-tuning

Abstract

Modern cameras' performance in low-light conditions remains suboptimal due to fundamental limitations in photon shot noise and sensor read noise. Generative image restoration methods have shown promising results compared to traditional approaches, but they suffer from hallucinatory content generation when the signal-to-noise ratio (SNR) is low. Leveraging the availability of personalized photo galleries of the users, we introduce Diffusion-based Personalized Generative Denoising (DiffPGD), a new approach that builds a customized diffusion model for individual users. Our key innovation lies in the development of an identity-consistent physical buffer that extracts the physical attributes of the person from the gallery. This ID-consistent physical buffer serves as a robust prior that can be seamlessly integrated into the diffusion model to restore degraded images without the need for fine-tuning. Over a wide range of low-light testing scenarios, we show that DiffPGD achieves superior image denoising and enhancement performance compared to existing diffusion-based denoising approaches. Our project page can be found at \href{https://genai-restore.github.io/DiffPGD/}{\textcolor{purple}{\textbf{https://genai-restore.github.io/DiffPGD/}}}.