Residual Feature Distillation Network for Lightweight Image Super-Resolution

TL;DR

This paper introduces the Residual Feature Distillation Network (RFDN), a lightweight CNN model for image super-resolution that balances high performance with low computational complexity, suitable for edge devices.

Contribution

It proposes the feature distillation connection (FDC) as a lightweight alternative to channel splitting, and develops RFDN and its enhanced version E-RFDN, achieving state-of-the-art results in efficient super-resolution.

Findings

RFDN outperforms existing methods in accuracy and efficiency.

FDC simplifies the model while maintaining performance.

E-RFDN won first place in the AIM 2020 super-resolution challenge.

Abstract

Recent advances in single image super-resolution (SISR) explored the power of convolutional neural network (CNN) to achieve a better performance. Despite the great success of CNN-based methods, it is not easy to apply these methods to edge devices due to the requirement of heavy computation. To solve this problem, various fast and lightweight CNN models have been proposed. The information distillation network is one of the state-of-the-art methods, which adopts the channel splitting operation to extract distilled features. However, it is not clear enough how this operation helps in the design of efficient SISR models. In this paper, we propose the feature distillation connection (FDC) that is functionally equivalent to the channel splitting operation while being more lightweight and flexible. Thanks to FDC, we can rethink the information multi-distillation network (IMDN) and propose a lightweight and accurate SISR model called residual feature distillation network (RFDN). RFDN uses multiple feature distillation connections to learn more discriminative feature representations. We also propose a shallow residual block (SRB) as the main building block of RFDN so that the network can benefit most from residual learning while still being lightweight enough. Extensive experimental results show that the proposed RFDN achieve a better trade-off against the state-of-the-art methods in terms of performance and model complexity. Moreover, we propose an enhanced RFDN (E-RFDN) and won the first place in the AIM 2020 efficient super-resolution challenge. Code will be available at https://github.com/njulj/RFDN.