Large Kernel Distillation Network for Efficient Single Image Super-Resolution
This work addresses computational efficiency for single-image super-resolution applications, but it appears incremental as it builds on existing large kernel designs with optimizations.
The paper tackles the problem of high computational costs in lightweight single-image super-resolution by proposing the Large Kernel Distillation Network, which simplifies the model structure and introduces efficient attention modules, resulting in state-of-the-art performance.
Efficient and lightweight single-image super-resolution (SISR) has achieved remarkable performance in recent years. One effective approach is the use of large kernel designs, which have been shown to improve the performance of SISR models while reducing their computational requirements. However, current state-of-the-art (SOTA) models still face problems such as high computational costs. To address these issues, we propose the Large Kernel Distillation Network (LKDN) in this paper. Our approach simplifies the model structure and introduces more efficient attention modules to reduce computational costs while also improving performance. Specifically, we employ the reparameterization technique to enhance model performance without adding extra cost. We also introduce a new optimizer from other tasks to SISR, which improves training speed and performance. Our experimental results demonstrate that LKDN outperforms existing lightweight SR methods and achieves SOTA performance.