I2V: Towards Texture-Aware Self-Supervised Blind Denoising using Self-Residual Learning for Real-World Images

TL;DR

This paper introduces a texture-preserving self-residual learning method for real-world blind denoising that outperforms existing approaches by avoiding the texture degradation caused by traditional downsampling techniques.

Contribution

The authors propose a novel self-residual learning approach that eliminates the need for pixel-shuffle downsampling, preserving texture details and improving denoising performance on real-world images.

Findings

Outperforms state-of-the-art self-supervised denoising methods in PSNR, SSIM, LPIPS, and DISTS.

Maintains high-frequency texture details better than PD-based methods.

Effective in real-world sRGB image denoising scenarios.

Abstract

Although the advances of self-supervised blind denoising are significantly superior to conventional approaches without clean supervision in synthetic noise scenarios, it shows poor quality in real-world images due to spatially correlated noise corruption. Recently, pixel-shuffle downsampling (PD) has been proposed to eliminate the spatial correlation of noise. A study combining a blind spot network (BSN) and asymmetric PD (AP) successfully demonstrated that self-supervised blind denoising is applicable to real-world noisy images. However, PD-based inference may degrade texture details in the testing phase because high-frequency details (e.g., edges) are destroyed in the downsampled images. To avoid such an issue, we propose self-residual learning without the PD process to maintain texture information. We also propose an order-variant PD constraint, noise prior loss, and an efficient inference scheme (progressive random-replacing refinement ($\text{PR}^3$)) to boost overall performance. The results of extensive experiments show that the proposed method outperforms state-of-the-art self-supervised blind denoising approaches, including several supervised learning methods, in terms of PSNR, SSIM, LPIPS, and DISTS in real-world sRGB images.