Learning to Translate Noise for Robust Image Denoising
This addresses the problem of robust image denoising for real-world applications, representing a novel method for a known bottleneck rather than incremental.
The paper tackles poor generalization of deep learning-based image denoising to out-of-distribution real-world noise by proposing a noise translation framework that converts complex noise into Gaussian noise, then uses a pretrained denoiser, resulting in substantial robustness improvements and outperforming state-of-the-art methods across benchmarks.
Deep learning-based image denoising techniques often struggle with poor generalization performance to out-of-distribution real-world noise. To tackle this challenge, we propose a novel noise translation framework that performs denoising on an image with translated noise rather than directly denoising an original noisy image. Specifically, our approach translates complex, unknown real-world noise into Gaussian noise, which is spatially uncorrelated and independent of image content, through a noise translation network. The translated noisy images are then processed by an image denoising network pretrained to effectively remove Gaussian noise, enabling robust and consistent denoising performance. We also design well-motivated loss functions and architectures for the noise translation network by leveraging the mathematical properties of Gaussian noise. Experimental results demonstrate that the proposed method substantially improves robustness and generalizability, outperforming state-of-the-art methods across diverse benchmarks. Visualized denoising results and the source code are available on our project page.