Nankai University

Overview: 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 high-frequency information under constrained resources.

This repository contains PyTorch implementation for EFDN (CVPRW 2022).

Run test_demo.py to reproduce results in ESR Challenge on NTIRE 2022 Workshop and Challenge.

The code of EDBB is based on ECBSR and DBB, and PAN Framework is utilized to train our EFDN. Thanks for their excellent work!

If you find our work useful in your research or publication, please cite our work:

@InProceedings{Wang_2022_CVPR,
 author = {Wang, Yan},
 title = {Edge-Enhanced Feature Distillation Network for Efficient Super-Resolution},
 booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
 month = {June},
 year = {2022},
 pages = {777-785}
}