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

2025-06-04

AI Summary

Existing blind raw image denoising methods heavily rely on real-world noise data from specific cameras, exhibiting poor generalization to unseen devices. This paper proposes YOND—the first blind denoising framework that achieves cross-camera generalization using only synthetic training data, without requiring any real camera-specific noise samples. Its key innovations include: (1) a coarse-to-fine noise estimation mechanism; (2) an expectation-matching variance-stabilizing transform (VST) that mitigates distributional discrepancies between synthetic and real noise; and (3) an SNR-guided multi-scale denoising network. By eliminating camera-specific modeling, YOND supports both manual tuning and adaptive inference. Evaluated on real raw images from diverse unknown cameras, YOND achieves state-of-the-art performance, significantly outperforming existing blind denoising approaches. It offers strong generalizability, controllability, and plug-and-play practicality.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Intelligent Robots: State Estimation

Application Category

• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Privacy-enhancing technologies

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}.

Problem

• Blind raw image denoising without camera-specific data dependency
• Generalizing denoising to diverse unknown cameras using synthetic data
• Overcoming performance drops in learning-based denoising methods

Innovation

• Coarse-to-fine noise estimation (CNE)
• Expectation-matched variance-stabilizing transform (EM-VST)
• SNR-guided denoiser (SNR-Net)