An implementation of FSRCNN in PyTorch

Originally authored by icpm

I did some minor modification to acquire better approach to the paper.

Requirement

dataset: create dir './dataset/BSDS500/images/[train, test]' and fill in your data

train:

$ python3 main.py

super resolve:

$ python3 super_resolve

Name		Name	Last commit message	Last commit date
Latest commit History 4 Commits
FSRCNN		FSRCNN
LICENSE		LICENSE
README.md		README.md
main.py		main.py
model_path.pth		model_path.pth
progress_bar.py		progress_bar.py
super_resolve.py		super_resolve.py