mlsanityData to 14 Oct 2026

Game State and Spatio-temporal Action Detection in Soccer using Graph Neural Networks and 3D Convolutional Networks

arXiv 2502.15462 ↗First posted 21 Feb 2025cs.CVcited by 6

Not on the map. Why?

Abstract (arXiv)

Soccer analytics rely on two data sources: the player positions on the pitch and the sequences of events they perform. With around 2000 ball events per game, their precise and exhaustive annotation based on a monocular video stream remains a tedious and costly manual task. While state-of-the-art spatio-temporal action detection methods show promise for automating this task, they lack contextual understanding of the game. Assuming professional players' behaviors are interdependent, we hypothesize that incorporating surrounding players' information such as positions, velocity and team membership can enhance purely visual predictions. We propose a spatio-temporal action detection approach that combines visual and game state information via Graph Neural Networks trained end-to-end with state-of-the-art 3D CNNs, demonstrating improved metrics through game state integration.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 35 of its 47 references that are in the map. First posted on arXiv 21 Feb 2025.

Builds on

35 of 47 references.

References · 35

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2023Visual Instruction Tuningcited by 7,869
  3. 2017Mask R-CNNcited by 7,277
  4. 2016Feature Pyramid Networks for Object Detectioncited by 5,036
  5. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,852
  6. 2024Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolutioncited by 4,234
  7. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,223
  8. 2018SlowFast Networks for Video Recognitioncited by 1,867
  9. 2020X3D: Expanding Architectures for Efficient Video Recognitioncited by 637
  10. 2017AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actionscited by 449
  11. 2018Video Action Transformer Networkcited by 316
  12. 2024Pixtral 12Bcited by 179
23 more
  1. 2020Actor-Transformers for Group Activity Recognitioncited by 84
  2. 2021MultiSports: A Multi-Person Video Dataset of Spatio-Temporally Localized Sports Actionscited by 84
  3. 2017Action Tubelet Detector for Spatio-Temporal Action Localizationcited by 77
  4. 2019You Only Watch Once: A Unified CNN Architecture for Real-Time Spatiotemporal Action Localizationcited by 59
  5. 2016Online Real-time Multiple Spatiotemporal Action Localisation and Predictioncited by 57
  6. 2020Actions as Moving Pointscited by 55
  7. 2021TubeR: Tubelet Transformer for Video Action Detectioncited by 55
  8. 2015Learning to track for spatio-temporal action localizationcited by 44
  9. 2016Social Scene Understanding: End-to-End Multi-Person Action Localization and Collective Activity Recognitioncited by 43
  10. 2017Artificial Intelligence and Statisticscited by 43
  11. 2018A Structured Model For Action Detectioncited by 43
  12. 2022Spotting Temporally Precise, Fine-Grained Events in Videocited by 40
  13. 2016Deep Learning for Detecting Multiple Space-Time Action Tubes in Videoscited by 35
  14. 2019Convolutional Relational Machine for Group Activity Recognitioncited by 34
  15. 2023SoccerNet 2023 Challenges Resultscited by 20
  16. 2018A Better Baseline for AVAcited by 16
  17. 2022Spatio-Temporal Action Detection Under Large Motioncited by 14
  18. 2022A Graph-Based Method for Soccer Action Spotting Using Unsupervised Player Classificationcited by 10
  19. 2023Inferring Player Location in Sports Matches: Multi-Agent Spatial Imputation from Limited Observationscited by 9
  20. 2021Agent-Environment Network for Temporal Action Proposal Generationcited by 8
  21. 2020A Graph Attention Based Approach for Trajectory Prediction in Multi-agent Sports Gamescited by 6
  22. 2022Video Analytics in Elite Soccer: A Distributed Computing Perspectivecited by 1
  23. 2024The Evolution of Football Betting- A Machine Learning Approach to Match Outcome Forecasting and Bookmaker Odds Estimationcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 3.

6 papers.

  1. 2025Beyond Pixels: Leveraging the Language of Soccer to Improve Spatio-Temporal Action Detection in Broadcast Videoscited by 5
  2. 2025FOOTPASS: A Multi-Modal Multi-Agent Tactical Context Dataset for Play-by-Play Action Spotting in Soccer Broadcast Videoscited by 5
  3. 2025Multi-Modal Soccer Scene Analysis with Masked Pre-Trainingcited by 1
  4. 2026SoccerNet 2026 Challenges Resultscited by 1
  5. 2026SoccerNet 2026 Player-Centric Ball-Action Spotting:Retraining and Post-Processing Extensions to the FOOTPASS Baselinescited by 0
  6. 2026Entity-Aware Sequence Transduction for Player-Centric Ball Action Spottingcited by 0