Pytorch code for Photon Limited Non-Blind Deblurring Using Algorithm Unrolling - published at Transactions on Computational Imaging
Pretrained model here
-
Create a local copy of repository using the following commands
foor@bar:~$ git clone https://github.com/sanghviyashiitb/poisson-deblurring.git foor@bar:~$ cd poisson-deblurring foor@bar:~/poisson-deblurring$
-
Download the pretrained model into
model_zoo
from the link here -
To test the network using synthetic data, run the file
foo@bar:~/poisson-deblurring$ python3 demo_synthetic.py
-
Download the zip file containing real dataset into the main directory and unzip using the following command:
foo@bar:~/poisson-deblurring$ unzip real_data.zip -d data/
-
To test the network using real data, run the file
foo@bar:~/poisson-deblurring$ python3 demo_synthetic.py --idx=11
(Variable
idx
represents the file index and can be any integer from [0,29] )
Before running train.py
, add clean images (for example Flickr2K) in the data/training
and data/val
folders.
@ARTICLE{9903556,
author={Sanghvi, Yash and Gnanasambandam, Abhiram and Chan, Stanley H.},
journal={IEEE Transactions on Computational Imaging},
title={Photon Limited Non-Blind Deblurring Using Algorithm Unrolling},
year={2022},
volume={8},
number={},
pages={851-864},
doi={10.1109/TCI.2022.3209939}}
Feel free to ask your questions/share your feedback at sanghviyash95@gmail.com