rucv/golf_ball

Efficient Golf Ball Detection and Tracking Based on Convolutional Neural Networks and Kalman Filter

We borrow the codes and implementations from jwyang-faster-rcnn.pytorch, and the Pytorch version is 1.0. Please refer to jwyang-faster-rcnn.pytorch for more details of setups and implementations.

Approach

Below is the framework of proposed method:

Dataset

The dataset link is golf_ball.

Results

Some tracking results are shown below:

Citation

@article{zhang2020efficient,
  title={Efficient Golf Ball Detection and Tracking Based on Convolutional Neural Networks and Kalman Filter},
  author={Zhang, Tianxiao and Zhang, Xiaohan and Yang, Yiju and Wang, Zongbo and Wang, Guanghui},
  journal={arXiv preprint arXiv:2012.09393},
  year={2020}
}


@inproceedings{zhang2020real,
  title={Real-time golf ball detection and tracking based on convolutional neural networks},
  author={Zhang, Xiaohan and Zhang, Tianxiao and Yang, Yiju and Wang, Zongbo and Wang, Guanghui},
  booktitle={2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC)},
  pages={2808--2813},
  year={2020},
  organization={IEEE}
}