Blind Image Super-Resolution With Efficient Network Design Using Frequency Domain Information

Abstract

Blind Image Super-Resolution (BSR) tackles the challenge of enhancing image resolution that has been degraded by unknown kernels. Although existing BSR models have achieved remarkable results, they often demand significant computational resources to manage various degradation kernels. Recent studies have utilized large models with parameter counts ranging from 4 M to 20 M, resulting in computational costs exceeding 200 G Multi-Adds. In this paper, we introduce a blind super-resolution method, which is the first lightweight non-iterative approach for BSR. By leveraging the connection between degradation kernel shapes and the frequency-domain characteristics of low-resolution images, we simplify the kernel estimation process, thereby reducing the overall model complexity. Additionally, we employ a constrained least squares approach to refine the low-resolution image using the estimated kernel, which serves as the input to the hierarchical Transformer blocks. Our approach delivers competitive performance while requiring only 1.7 M parameters and 2 G Multi-Adds. Experimental results demonstrate that our method achieves comparable results with significantly smaller network size.