Efficient Attention-Sharing Information Distillation Transformer for Lightweight Single Image Super-Resolution
This addresses the efficiency bottleneck for practical deployment of lightweight single image super-resolution models, though it is incremental as it adapts existing techniques to Transformers.
The paper tackles the high computational complexity of Transformer-based super-resolution methods by proposing the Attention-Sharing Information Distillation (ASID) network, which achieves competitive performance with only around 300K parameters and outperforms state-of-the-art methods when parameters are matched.
Transformer-based Super-Resolution (SR) methods have demonstrated superior performance compared to convolutional neural network (CNN)-based SR approaches due to their capability to capture long-range dependencies. However, their high computational complexity necessitates the development of lightweight approaches for practical use. To address this challenge, we propose the Attention-Sharing Information Distillation (ASID) network, a lightweight SR network that integrates attention-sharing and an information distillation structure specifically designed for Transformer-based SR methods. We modify the information distillation scheme, originally designed for efficient CNN operations, to reduce the computational load of stacked self-attention layers, effectively addressing the efficiency bottleneck. Additionally, we introduce attention-sharing across blocks to further minimize the computational cost of self-attention operations. By combining these strategies, ASID achieves competitive performance with existing SR methods while requiring only around 300K parameters - significantly fewer than existing CNN-based and Transformer-based SR models. Furthermore, ASID outperforms state-of-the-art SR methods when the number of parameters is matched, demonstrating its efficiency and effectiveness. The code and supplementary material are available on the project page.