Object Tracking in Satellite Videos by Improved Correlation Filters With Motion Estimations

This repo contains the source code for our TGRS work Object Tracking in Satellite Videos by Improved Correlation Filters With Motion Estimations. This repo is based on the code https://github.com/uoip/KCFpy.

Prerequisites

1. Python3
2. Opencv
3. Numpy
4. Numba

Run the demo

you need to modify some file path in the demo.py

    python3 demo.py

The File Structure

The file needs to be look like

├── 00001.jpg
├── 00002.jpg
|── 00003.jpg

Refer to this Rep.

If you found our work is useful, thanks to you to cite our paper and star.

@ARTICLE{8880656,
  author={S. {Xuan} and S. {Li} and M. {Han} and X. {Wan} and G. {Xia}},
  journal={IEEE Transactions on Geoscience and Remote Sensing}, 
  title={Object Tracking in Satellite Videos by Improved Correlation Filters With Motion Estimations}, 
  year={2020},
  volume={58},
  number={2},
  pages={1074-1086},
  doi={10.1109/TGRS.2019.2943366}}

Contact

If you have any problem feel free to email me at xuanshiyu17@csu.ac.cn.