Home

Awesome

RS-CLIP: Zero Shot Remote Sensing Scene Classification via Contrastive Vision-Language Supervision

Introduction

overview

Getting Started

Installation

Following CLIP to install required packages

Reproduce

refer to run_ucm.sh for experiments on the UCM dataest.

Acknowledgement

If you're using RS-CLIP in your research or applications, please cite using this BibTeX:

@article{li2023rs,
  title={RS-CLIP: Zero shot remote sensing scene classification via contrastive vision-language supervision},
  author={Li, Xiang and Wen, Congcong and Hu, Yuan and Zhou, Nan},
  journal={International Journal of Applied Earth Observation and Geoinformation},
  volume={124},
  pages={103497},
  year={2023},
  publisher={Elsevier}
}

License

This repository is under BSD 3-Clause License. BSD 3-Clause License here.