Edge-enhanced Feature Distillation Network for Efficient Super-Resolution

TL;DR

This paper introduces EFDN, an efficient super-resolution network that enhances edge preservation and high-frequency details using edge-enhanced convolution and gradient loss, suitable for resource-constrained devices.

Contribution

The paper proposes an edge-enhanced feature distillation network with novel edge-enhanced convolution blocks and gradient loss for improved super-resolution performance.

Findings

Significantly improves edge preservation in super-resolution.

Enhances restoration quality with edge-enhanced strategies.

Achieves better performance under resource constraints.

Abstract

With the recently massive development in convolution neural networks, numerous lightweight CNN-based image super-resolution methods have been proposed for practical deployments on edge devices. However, most existing methods focus on one specific aspect: network or loss design, which leads to the difficulty of minimizing the model size. To address the issue, we conclude block devising, architecture searching, and loss design to obtain a more efficient SR structure. In this paper, we proposed an edge-enhanced feature distillation network, named EFDN, to preserve the high-frequency information under constrained resources. In detail, we build an edge-enhanced convolution block based on the existing reparameterization methods. Meanwhile, we propose edge-enhanced gradient loss to calibrate the reparameterized path training. Experimental results show that our edge-enhanced strategies preserve the edge and significantly improve the final restoration quality. Code is available at https://github.com/icandle/EFDN.