Skip to content

A SoccerNet v3 subset for Person and Ball detection in YOLO format

License

MIT, MIT licenses found

Licenses found

MIT
LICENSE
MIT
LICENSE_SoccerNetv3
Notifications You must be signed in to change notification settings

kmouts/SoccerNet_v3_H250

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 

Repository files navigation

SoccerNet v3 H250 (YOLO annotations for Ball & Player Detection)

Setting a Baseline for long-shot real-time Player and Ball detection in Soccer Videos](https://easyconferences.eu/iisa2023/) IISA 2023 Volos, Greece.

Created by Kostas Moutselos, and Ilias Maglogiannis at CBML

football match detection image

Abstract

Players and ball detection are among the first required steps on a football analytics platform. Until recently, the existing open datasets on which the evaluations of most models were based, were not sufficient. We point out their weaknesses, and with the advent of SoccerNet v3, we propose and deliver to the community an edited part of it, in YOLO normalized annotation format, for the reliable evaluation of related models https://doi.org/10.5281/zenodo.7808511. The code of the methods and metrics are also available so that they can be used as a benchmark in future comparisons. [https://github.com/kmouts/FootAndBall]. The recent YOLO8n model proves better than FootAndBall in long-shot real-time detection of the ball and players on football fields.

Directory Structure

Directory Structure

YOLO Normalized Annotations

YOLO bboxes annotation format

Annotations

Only two classes are included in the YOLO annotations: Ball (0), and Person (1).

Dataset Statistics

Daset statistics regarding the train (green), validation (orange) and test (red) parts. Dataset Statistics

Citation

Please, if you use the SoccerNet v3 H250 Dataset, consider citing:

@INPROCEEDINGS{10345947,
author={Moutselos, Konstantinos and Maglogiannis, Ilias},
booktitle={2023 14th International Conference on Information, Intelligence, Systems & Applications (IISA)}, 
title={Setting a Baseline for Long-Shot Real-Time Player and Ball Detection in Soccer Videos}, 
year={2023},
volume={},
number={},
pages={1-7},
doi={10.1109/IISA59645.2023.10345947}}
}

License

Our code is released under the MIT License (see LICENSE file for details). We have added the License from SoccerNet v3.

About

A SoccerNet v3 subset for Person and Ball detection in YOLO format

Resources

License

MIT, MIT licenses found

Licenses found

MIT
LICENSE
MIT
LICENSE_SoccerNetv3

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published