Reconstruct-and-Generate Diffusion Model for Detail-Preserving Image Denoising

TL;DR

This paper introduces the Reconstruct-and-Generate Diffusion Model (RnG) for image denoising, which combines a reconstructive network with a diffusion process to enhance detail preservation and visual quality while controlling artifacts.

Contribution

The paper proposes a novel two-stage RnG framework that integrates reconstructive denoising with diffusion-based residual detail generation, including an adaptive step controller for better quality and efficiency.

Findings

Outperforms existing denoising methods in preserving details

Effectively balances visual quality and fidelity

Reduces artifacts with adaptive diffusion steps

Abstract

Image denoising is a fundamental and challenging task in the field of computer vision. Most supervised denoising methods learn to reconstruct clean images from noisy inputs, which have intrinsic spectral bias and tend to produce over-smoothed and blurry images. Recently, researchers have explored diffusion models to generate high-frequency details in image restoration tasks, but these models do not guarantee that the generated texture aligns with real images, leading to undesirable artifacts. To address the trade-off between visual appeal and fidelity of high-frequency details in denoising tasks, we propose a novel approach called the Reconstruct-and-Generate Diffusion Model (RnG). Our method leverages a reconstructive denoising network to recover the majority of the underlying clean signal, which serves as the initial estimation for subsequent steps to maintain fidelity. Additionally, it employs a diffusion algorithm to generate residual high-frequency details, thereby enhancing visual quality. We further introduce a two-stage training scheme to ensure effective collaboration between the reconstructive and generative modules of RnG. To reduce undesirable texture introduced by the diffusion model, we also propose an adaptive step controller that regulates the number of inverse steps applied by the diffusion model, allowing control over the level of high-frequency details added to each patch as well as saving the inference computational cost. Through our proposed RnG, we achieve a better balance between perception and distortion. We conducted extensive experiments on both synthetic and real denoising datasets, validating the superiority of the proposed approach.