LKFMixer: Exploring Large Kernel Feature For Efficient Image Super-Resolution

TL;DR

LKFMixer is a convolutional neural network that uses large kernels and novel modules to efficiently capture non-local features for superior image super-resolution, outperforming state-of-the-art methods in quality and speed.

Contribution

The paper introduces LKFMixer, a CNN with large kernels and new modules that simulate self-attention for efficient super-resolution without heavy computation.

Findings

Outperforms SOTA methods in SR quality and reconstruction.

Achieves 0.6dB PSNR improvement over SwinIR-light on Manga109.

Runs 5 times faster than comparable models.

Abstract

The success of self-attention (SA) in Transformer demonstrates the importance of non-local information to image super-resolution (SR), but the huge computing power required makes it difficult to implement lightweight models. To solve this problem, we propose a pure convolutional neural network (CNN) model, LKFMixer, which utilizes large convolutional kernel to simulate the ability of self-attention to capture non-local features. Specifically, we increase the kernel size to 31 to obtain the larger receptive field as possible, and reduce the parameters and computations by coordinate decomposition. Meanwhile, a spatial feature modulation block (SFMB) is designed to enhance the focus of feature information on both spatial and channel dimension. In addition, by introducing feature selection block (FSB), the model can adaptively adjust the weights between local features and non-local features. Extensive experiments show that the proposed LKFMixer family outperform other state-of-the-art (SOTA) methods in terms of SR performance and reconstruction quality. In particular, compared with SwinIR-light on Manga109 dataset, LKFMixer-L achieves 0.6dB PSNR improvement at $\times$4 scale, while the inference speed is $\times$5 times faster. The code is available at https://github.com/Supereeeee/LKFMixer.