Learning to Translate Noise for Robust Image Denoising
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917378996043776 |
|---|---|
| author | Ha, Inju Ryou, Donghun Seo, Seonguk Han, Bohyung |
| author_facet | Ha, Inju Ryou, Donghun Seo, Seonguk Han, Bohyung |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_04727 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Learning to Translate Noise for Robust Image Denoising Ha, Inju Ryou, Donghun Seo, Seonguk Han, Bohyung Image and Video Processing Computer Vision and Pattern Recognition 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. |
| title | Learning to Translate Noise for Robust Image Denoising |
| topic | Image and Video Processing Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.04727 |