Stochastic Frequency Masking to Improve Super-Resolution and Denoising Networks
This addresses a key limitation in learning-based image restoration methods for practical applications where degradation parameters are unknown.
The paper tackles the problem of degradation-kernel overfitting in blind super-resolution and denoising networks by proposing stochastic frequency masking during training, which improves state-of-the-art methods on various synthetic and real-world benchmarks.
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.