TY - JOUR
T1 - YOND
T2 - Practical Blind Raw Image Denoising Free from Camera-Specific Data Dependency
AU - Feng, Hansen
AU - Wang, Lizhi
AU - Huang, Yiqi
AU - Li, Tong
AU - Zhu, Lin
AU - Huang, Hua
N1 - Publisher Copyright:
© 1979-2012 IEEE.
PY - 2026
Y1 - 2026
N2 - 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), expectationmatched variance-stabilizing transform (EM-VST), and SNRguided 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.
AB - 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), expectationmatched variance-stabilizing transform (EM-VST), and SNRguided 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.
KW - Blind Raw Image Denoising
KW - Computational Photography
KW - Data Dependency
KW - Diffusion Model
UR - https://www.scopus.com/pages/publications/105051572534
U2 - 10.1109/TPAMI.2026.3736494
DO - 10.1109/TPAMI.2026.3736494
M3 - Article
C2 - 42771615
AN - SCOPUS:105051572534
SN - 0162-8828
JO - IEEE Transactions on Pattern Analysis and Machine Intelligence
JF - IEEE Transactions on Pattern Analysis and Machine Intelligence
ER -