IDR: Self-Supervised Image Denoising via Iterative Data Refinement

TL;DR

This paper introduces IDR, a practical unsupervised image denoising method that iteratively refines noisy images using only single noisy images and a noise model, achieving state-of-the-art results without requiring clean image pairs.

Contribution

The authors propose a novel iterative data refinement approach for unsupervised denoising that works with only noisy images and a noise model, improving performance over existing methods.

Findings

Achieves superior denoising performance on real-world and synthetic data.

Introduces the SenseNoise-500 dataset for benchmarking raw image denoising.

Provides a fast algorithm approximation maintaining high performance.

Abstract

The lack of large-scale noisy-clean image pairs restricts supervised denoising methods' deployment in actual applications. While existing unsupervised methods are able to learn image denoising without ground-truth clean images, they either show poor performance or work under impractical settings (e.g., paired noisy images). In this paper, we present a practical unsupervised image denoising method to achieve state-of-the-art denoising performance. Our method only requires single noisy images and a noise model, which is easily accessible in practical raw image denoising. It performs two steps iteratively: (1) Constructing a noisier-noisy dataset with random noise from the noise model; (2) training a model on the noisier-noisy dataset and using the trained model to refine noisy images to obtain the targets used in the next round. We further approximate our full iterative method with a fast algorithm for more efficient training while keeping its original high performance. Experiments on real-world, synthetic, and correlated noise show that our proposed unsupervised denoising approach has superior performances over existing unsupervised methods and competitive performance with supervised methods. In addition, we argue that existing denoising datasets are of low quality and contain only a small number of scenes. To evaluate raw image denoising performance in real-world applications, we build a high-quality raw image dataset SenseNoise-500 that contains 500 real-life scenes. The dataset can serve as a strong benchmark for better evaluating raw image denoising. Code and dataset will be released at https://github.com/zhangyi-3/IDR