Stochastic Frequency Masking to Improve Super-Resolution and Denoising Networks
Abstract
Super-resolution and denoising are ill-posed yet fundamental image restoration tasks. In blind settings, the degradation kernel or the noise level are unknown. This makes restoration even more challenging, notably for learning-based methods, as they tend to overfit to the degradation seen during training. We present an analysis, in the frequency domain, of degradation-kernel overfitting in super-resolution and introduce a conditional learning perspective that extends to both super-resolution and denoising. Building on our formulation, we propose a stochastic frequency masking of images used in training to regularize the networks and address the overfitting problem. Our technique improves state-of-the-art methods on blind super-resolution with different synthetic kernels, real super-resolution, blind Gaussian denoising, and real-image denoising.
Cite
Text
El Helou et al. "Stochastic Frequency Masking to Improve Super-Resolution and Denoising Networks." Proceedings of the European Conference on Computer Vision (ECCV), 2020. doi:10.1007/978-3-030-58517-4_44Markdown
[El Helou et al. "Stochastic Frequency Masking to Improve Super-Resolution and Denoising Networks." Proceedings of the European Conference on Computer Vision (ECCV), 2020.](https://mlanthology.org/eccv/2020/helou2020eccv-stochastic/) doi:10.1007/978-3-030-58517-4_44BibTeX
@inproceedings{helou2020eccv-stochastic,
title = {{Stochastic Frequency Masking to Improve Super-Resolution and Denoising Networks}},
author = {El Helou, Majed and Zhou, Ruofan and Süsstrunk, Sabine},
booktitle = {Proceedings of the European Conference on Computer Vision (ECCV)},
year = {2020},
doi = {10.1007/978-3-030-58517-4_44},
url = {https://mlanthology.org/eccv/2020/helou2020eccv-stochastic/}
}