maverick-yan/pgman

code for "PGMAN: An Unsupervised Generative Multi-adversarial Network for Pan-sharpening"

PGMAN

This repo is the official implementation for PGMAN: An Unsupervised Generative Multiadversarial Network for Pansharpening.

The paper is accepted to J-STARS2021.

implemented in PyTorch1.1

python lib see in requirement.txt

Architecture

Raw Data

link: https://pan.baidu.com/s/190MywbwIlvONA_9-6-KMtQ code: u041

link: https://pan.baidu.com/s/1dRrMH6KcFnkGuYZMCcJooA code: odg0

Quick Start

First download the raw data, and then build the dataset.

python data/handle_raw.py
python data/gen_dataset.py

It's suggested to check the corresponding path in the codes and make some modifications according to yourself before executing it.

The main pipeline is in the 'main.py', for a quick start, you can just run the 'run.py'.

python run.py

It's suggested to check the corresponding params in 'run.py' and make some modifications according to yourself before executing it.

Citing PGMAN

Consider cite PGMAN in your publications if it helps your research.

@ARTICLE{pgman,
  author={Zhou, Huanyu and Liu, Qingjie and Wang, Yunhong},
  journal={IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing}, 
  title={PGMAN: An Unsupervised Generative Multiadversarial Network for Pansharpening}, 
  year={2021},
  volume={14},
  number={},
  pages={6316-6327},
}