YOND: Practical Blind Raw Image Denoising Free from Camera-Specific Data Dependency

TL;DR

YOND is a practical blind raw image denoising method that generalizes well to unknown cameras by using synthetic training data and novel modules to eliminate camera-specific dependencies.

Contribution

The paper introduces YOND, a blind denoising approach with three key modules that enable robust performance across diverse cameras without camera-specific training.

Findings

YOND outperforms existing methods on unknown camera data.

The method effectively eliminates camera-specific data dependency.

YOND supports adaptive and manual fine-tuning for denoising quality.

Abstract

The rapid advancement of photography has created a growing demand for a practical blind raw image denoising method. Recently, learning-based methods have become mainstream due to their excellent performance. However, most existing learning-based methods suffer from camera-specific data dependency, resulting in performance drops when applied to data from unknown cameras. To address this challenge, we introduce a novel blind raw image denoising method named YOND, which represents You Only Need a Denoiser. Trained solely on synthetic data, YOND can generalize robustly to noisy raw images captured by diverse unknown cameras. Specifically, we propose three key modules to guarantee the practicality of YOND: coarse-to-fine noise estimation (CNE), expectation-matched variance-stabilizing transform (EM-VST), and SNR-guided denoiser (SNR-Net). Firstly, we propose CNE to identify the camera noise characteristic, refining the estimated noise parameters based on the coarse denoised image. Secondly, we propose EM-VST to eliminate camera-specific data dependency, correcting the bias expectation of VST according to the noisy image. Finally, we propose SNR-Net to offer controllable raw image denoising, supporting adaptive adjustments and manual fine-tuning. Extensive experiments on unknown cameras, along with flexible solutions for challenging cases, demonstrate the superior practicality of our method. The source code will be publicly available at the \href{https://fenghansen.github.io/publication/YOND}{project homepage}.