This code is part of the supplementary materials for the ISVC 2020 for our paper Fundus2Angio: A Conditional GAN Architecture for Generating Fluorescein Angiography Images from Retinal Fundus Photography . The paper has since been accpeted to ISVC 2020 and will be preseneted in October 2020.
https://arxiv.org/abs/2005.05267
@inproceedings{kamran2020fundus2angio,
title={Fundus2Angio: A Conditional GAN Architecture for Generating Fluorescein Angiography Images from Retinal Fundus Photography},
author={Kamran, Sharif Amit and Hossain, Khondker Fariha and Tavakkoli, Alireza and Zuckerbrod, Stewart and Baker, Salah A and Sanders, Kenton M},
booktitle={International Symposium on Visual Computing},
pages={125--138},
year={2020},
organization={Springer}
}
- Ubuntu 18.04 / Windows 7 or later
- NVIDIA Graphics card
- Download and Install Nvidia Drivers
- Download and Install via Runfile Nvidia Cuda Toolkit 10.0
- Download and Install Nvidia CuDNN 7.6.5 or later
- Install Pip3 and Python3 enviornment
sudo apt-get install pip3 python3-dev
- Install Tensorflow-Gpu version-2.0.0 and Keras version-2.3.1
sudo pip3 install tensorflow-gpu==2.0.3
sudo pip3 install keras==2.3.1
- Install packages from requirements.txt
sudo pip3 -r requirements.txt
https://sites.google.com/site/hosseinrabbanikhorasgani/datasets-1/fundus-fluorescein-angiogram-photographs--colour-fundus-images-of-diabetic-patients
- Please cite the paper if you use their data
@article{hajeb2012diabetic,
title={Diabetic retinopathy grading by digital curvelet transform},
author={Hajeb Mohammad Alipour, Shirin and Rabbani, Hossein and Akhlaghi, Mohammad Reza},
journal={Computational and mathematical methods in medicine},
volume={2012},
year={2012},
publisher={Hindawi}
}
- Folder structure for data pre-processing given below. Please make sure it matches with your local repository.
├── Dataset
| ├──ABNORMAL
| ├──NORMAL
- Type this in terminal to run the random_crop.py file
python3 random_crop.py --output_dir=data --input_dim=512 --datadir=Dataset
- There are different flags to choose from. Not all of them are mandatory.
'--input_dim', type=int, default=512
'--n_crops', type=int, default=50
'--datadir', type=str, required=True, help='path/to/data_directory',default='Dataset'
'--output_dir', type=str, default='data'
- Convert all the images to npz format
python3 convert_npz.py --outfile_name=fundus2angio --input_dim=512 --datadir=data --n_crops=50
- There are different flags to choose from. Not all of them are mandatory.
'--input_dim', type=int, default=512
'--n_crops', type=int, default=50
'--datadir', type=str, required=True, help='path/to/data_directory',default='data'
'--outfile_name', type=str, default='fundus2angio'
'--n_images', type=int, default=17
- Type this in terminal to run the train.py file
python3 train.py --npz_file=fundus2angio --batch=4 --epochs=100
- There are different flags to choose from. Not all of them are mandatory
'--npz_file', type=str, default='fundus2angio', help='path/to/npz/file'
'--batch_size', type=int, default=4
'--input_dim', type=int, default=512
'--epochs', type=int, default=100
The code is released under the MIT License, you can read the license file included in the repository for details.
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent