The Challenge Awakens: SoccerNet 2025!

Описание: Join the SoccerNet community and take part in the 2025 challenges in soccer video analysis. Be part of the research community pushing the boundaries of broadcast, field, and player understanding. With new tasks and challenges, including multi-view foul recognition, game state reconstruction, team ball action spotting, and monocular depth estimation, there's a challenge for everyone. With prizes for each challenge and a chance to be recognized at CVPR, start your journey now and help shape the future of soccer video analysis. Join our Discord community, follow us on social media, and be sure to leave a like and subscribe to our channel for updates. Let's get started!

00:00​​ Introduction
01:34 Multi-view Foul Recognition challenge
03:06 Game State Reconstruction challenge
05:33 Team Ball Action Spotting challenge
08:03 Monocular Depth Estimation challenge
09:32 Conclusion

Make sure to join our SoccerNet community on Discord:   / discord  

Link to our website: https://www.soccer-net.org

Multi-view Foul Recognition Challenge
Link to the repository: https://github.com/SoccerNet/sn-mvfoul
Link to the VARS paper: https://openaccess.thecvf.com/content...
Sponsor: TBD
Prize: TBD
Venue: CVsports workshop at CVPR

Game state reconstruction Challenge
Link to the repository: https://github.com/SoccerNet/sn-games...
Sponsor: TBD
Prize: TBD
Venue: CVSports workshop at CVPR

Team Ball Action Spotting Challenge
Link to the repository: https://github.com/SoccerNet/sn-teams...
Sponsor: TBD
Prizes: TBD
Venue: CVsports workshop at CVPR

Monocular Depth Estimation Challenge
Link to the repository: https://github.com/SoccerNet/sn-depth
Sponsor: TBD
Prizes: TBD
Venue: CVsports workshop at CVPR


Big thanks to Footovision (https://www.footovision.com/) for providing the Ball Action Spotting data and annotations and to Baidu Research for their original contribution to the SoccerNet Tracking dataset at the core of the Game State task!

Pip package to get started with the dataset and the challenge:
https://pypi.org/project/SoccerNet/​​​

Previous tutorials to get you started
• Tutorial #1    • Tutorial #1 Introduction to the 2022 Socce...  
• Tutorial #2    • Tutorial #2 Action Spotting & Replay Groun...  
• Tutorial #3    • Tutorial #3 Camera Calibration - 2022 Socc...  
• Tutorial #4    • Tutorial #4 Player Re-Identification - 202...  
• Tutorial #5    • Tutorial #5 Player Tracking - 2022 SoccerN...  
• Tutorial #6    • SoccerNet 2023 Live Tutorials - ft. Vladim...  
• Tutorial #7    • SoccerNet 2024 Live Tutorials - ft. Vladim...  

Don’t hesitate to contact us if you have any questions about the papers, the challenges or the code!


To see more of our work, subscribe to our YouTube channel and other social media!
We’ll keep posting our work regularly.

Subscribe to Acad Research:
   / acadresearch  

Twitter:
  / soccernet_org  
  / anthony_cioppa  
  / adriendeliege  
  / giancolasilvio  
  / baiduresearch  
  / _cvsports  

Other social media:
  / vladimirsomers  


Video Credits
-------
Background videos from Videezy.com and Pexels
VAR Room - AFP

Music credits
-------
Juno in The Space Maze - Loopop
Stranger - VYEN
Moons - Patrick Patrikios
Copycat - R.LUM.R
Missive - Andrew Langdon
Demon - JVNA
Classic - Joakim Karud
-------

Many thanks to the University of Liège, KAUST, UCLouvain, EPFL, Aalborg University, Footovision, TRAIL, FNRS, Sportradar, Baidu Research, NASK Science, UPC, University of Barcelone, and EVS Broadcast Equipment.

Written and Presented by
Anthony Cioppa - ULiège
Jan Held - ULiège & KAUST
Arnaud Leduc - ULiège
Silvio Giancola - KAUST
Vladimir Somers - UCLouvain, EPFL, and Sportradar
Victor Joos de ter Beerst - UCLouvain
Seyed Abolfazl Ghasemzadeh - UCLouvain
Marc Gutiérrez - UPC
Artur Xarles - University of Barcelona

Produced and Edited by Anthony Cioppa.