This implementation follows the approach presented in [1].
Furthermore, the approach was validated against the predictions of a pre-trained YOLOv5. Results can be found in the report.pdf
.
I strongly recommend to those getting started with Kalman filters to give [2] a chance. Great quality material.
[1] Kim, J. Y., & Kim, T. Y. (2009, August). Soccer ball tracking using dynamic kalman filter with velocity control. In 2009 Sixth International Conference on computer graphics, imaging and visualization (pp. 367-374). IEEE.
[2] Kalman-and-Bayesian-Filters-in-Python. https://github.com/rlabbe/Kalman-and-Bayesian-Filters-in-Python/tree/master/