KPAC-Net (Retrained Version)

This repository offers the retrained version of KPAC network using the training strategy mentioned in the following paper.

Learning to Deblur using Light Field Generated and Real Defocused Images
Lingyan Ruan^*, Bin Chen^*, Jizhou Li, Miuling Lam （* equal contribution）
IEEE Computer Vision and Pattern Recognition (CVPR Oral) 2022

The orignial KPAC-Net can be found Here .

Code Here

Prerequisites

Notes: the code may also work with other library versions that didn't specify here.

Table 4 in our main paper.

1. Installation

Clone this project to your local machine

$ git clone https://github.com/lingyanruan/KPAC-Net.git
$ cd KPAC-Net

2. Pre-trained models

Download and unzip pretrained weights under ./ModelCheckpoints/:

3. Datasets

Download and unzip test sets DPDD under folder ./DPD/

4. Command Line

python Eval.py

Contact

Should you have any questions, please open an issue or contact me lyruanruan@gmail.com

License

This software is being made available under the terms in the LICENSE file.