Action Spotting in Soccer Videos Using Dense Detection Anchors

This repository contains models for action spotting in soccer videos. For more details around the models and instructions on how to use them, please see https://github.com/yahoo/spivak.

The technical details of our models are described in the following two papers, which focus on action spotting on the SoccerNet dataset.

Temporally Precise Action Spotting in Soccer Videos Using Dense Detection Anchors. In ICIP, 2022.
Action Spotting using Dense Detection Anchors Revisited: Submission to the SoccerNet Challenge 2022. arXiv preprint, 2022.

Our action spotting models came in first place in the SoccerNet Challenge 2022. You can read more about the 2022 challenge and results in the following paper:

SoccerNet 2022 Challenges Results. In MMSports, 2022.

Citations

If you found our models and code useful, please consider citing our works:

@inproceedings{soares2022temporally,
  author={Soares, Jo{\~a}o~V.~B. and Shah, Avijit and Biswas, Topojoy},
  booktitle={International Conference on Image Processing (ICIP)},
  title={Temporally Precise Action Spotting in Soccer Videos Using Dense Detection Anchors},
  year={2022},
  pages={2796-2800},
  doi={10.1109/ICIP46576.2022.9897256}
}

@article{soares2022action,
  title={Action Spotting using Dense Detection Anchors Revisited: Submission to the {SoccerNet} {Challenge} 2022},
  author={Soares, Jo{\~a}o~V.~B. and Shah, Avijit},
  journal={arXiv preprint arXiv:2206.07846},
  year={2022}
}

Maintainers

Joao Soares: jvbsoares@yahooinc.com
Avijit Shah: avijit.shah@yahooinc.com

License

The contents of this repository are (c) by Yahoo Inc.

The contents of this repository are licensed under a Creative Commons Attribution 4.0 International License.

You should have received a copy of the license along with this work. If not, see https://creativecommons.org/licenses/by/4.0/.