https://github.com/mewil/res-gan

a residual GAN that removes clouds from satellite imagery
https://github.com/mewil/res-gan

gan satellite-imagery

Last synced: 4 days ago
JSON representation

a residual GAN that removes clouds from satellite imagery

• Host: GitHub
• URL: https://github.com/mewil/res-gan
• Owner: mewil
• Created: 2020-04-20T14:22:00.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2020-04-28T17:05:39.000Z (over 4 years ago)
• Last Synced: 2024-01-29T17:12:28.879Z (10 months ago)
• Topics: gan, satellite-imagery
• Language: Python
• Homepage:
• Size: 39.1 KB
• Stars: 1
• Watchers: 3
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Residual cGANs for Cloud Cover Removal

Final Project for University of Michigan EECS 442. Code adapted from [Enomoto et al](https://github.com/enomotokenji/mcgan-cvprw2017-pytorch) and [Python-Clouds](https://github.com/SquidDev/Python-Clouds).

Downloading and generating the dataset:

```sh

# AWS CLI must be installed

cd data/

mkdir images

# Download and unzip the data

aws s3 cp s3://spacenet-dataset/spacenet/SN1_buildings/tarballs/SN1_buildings_train_AOI_1_Rio_3band.tar.gz .

aws s3 cp s3://spacenet-dataset/spacenet/SN1_buildings/tarballs/SN1_buildings_test_AOI_1_Rio_3band.tar.gz .

tar -xvzf SN1_buildings_train_AOI_1_Rio_3band.tar.gz -C images/

tar -xvzf SN1_buildings_test_AOI_1_Rio_3band.tar.gz -C images/

rm SN1_buildings_train_AOI_1_Rio_3band.tar.gz

rm SN1_buildings_test_AOI_1_Rio_3band.tar.gz

# Then split the data into train, validation, and test directories before running the cloud generation script

python3 generate_clouds.py --input_dir train

python3 generate_clouds.py --input_dir validation

python3 generate_clouds.py --input_dir test

```