mlsanityData to 7 Oct 2026

A Context-Aware Loss Function for Action Spotting in Soccer Videos

arXiv 1912.01326 ↗First posted 3 Dec 2019cs.CV cs.LG eess.IVcited by 56

Not on the map. Why?

Abstract (arXiv)

In video understanding, action spotting consists in temporally localizing human-induced events annotated with single timestamps. In this paper, we propose a novel loss function that specifically considers the temporal context naturally present around each action, rather than focusing on the single annotated frame to spot. We benchmark our loss on a large dataset of soccer videos, SoccerNet, and achieve an improvement of 12.8% over the baseline. We show the generalization capability of our loss for generic activity proposals and detection on ActivityNet, by spotting the beginning and the end of each activity. Furthermore, we provide an extended ablation study and display challenging cases for action spotting in soccer videos. Finally, we qualitatively illustrate how our loss induces a precise temporal understanding of actions and show how such semantic knowledge can be used for automatic highlights generation.

Key facts. Cited by 56 papers in the map (data through 7 Oct 2026). First posted on arXiv 3 Dec 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 7; 2022: 13; 2023: 10; 2024: 10; 2025: 8; 2026: 8.

56 papers.

Sort:

By area:

  1. 2022SoccerNet-Tracking: Multiple Object Tracking Dataset and Benchmark in Soccer Videoscited by 80
  2. 2023SoccerNet-Caption: Dense Video Captioning for Soccer Broadcasts Commentariescited by 57
  3. 2021Camera Calibration and Player Localization in SoccerNet-v2 and Investigation of their Representations for Action Spottingcited by 43
  4. 2022Spotting Temporally Precise, Fine-Grained Events in Videocited by 40
  5. 2023VARS: Video Assistant Referee System for Automated Soccer Decision Making from Multiple Viewscited by 38
  6. 2024SoccerNet Game State Reconstruction: End-to-End Athlete Tracking and Identification on a Minimapcited by 34
  7. 2021Deep Learning-based Action Detection in Untrimmed Videos: A Surveycited by 32
  8. 2021Temporally-Aware Feature Pooling for Action Spotting in Soccer Broadcastscited by 30
  9. 2021Feature Combination Meets Attention: Baidu Soccer Embeddings and Transformer based Temporal Detectioncited by 27
  10. 2022Temporally Precise Action Spotting in Soccer Videos Using Dense Detection Anchorscited by 27
  11. 2022A Survey on Video Action Recognition in Sports: Datasets, Methods and Applicationscited by 27
  12. 2022MonoTrack: Shuttle trajectory reconstruction from monocular badminton videocited by 24
38 more
  1. 2024X-VARS: Introducing Explainability in Football Refereeing with Multi-Modal Large Language Modelcited by 22
  2. 2022A Comprehensive Review of Computer Vision in Sports: Open Issues, Future Trends and Research Directionscited by 21
  3. 2021RMS-Net: Regression and Masking for Soccer Event Spottingcited by 19
  4. 2023COMEDIAN: Self-Supervised Learning and Knowledge Distillation for Action Spotting using Transformerscited by 17
  5. 2024ASTRA: An Action Spotting TRAnsformer for Soccer Videoscited by 16
  6. 2022Action Spotting using Dense Detection Anchors Revisited: Submission to the SoccerNet Challenge 2022cited by 14
  7. 2023Towards Active Learning for Action Spotting in Association Football Videoscited by 14
  8. 2022Temporal Perceiver: A General Architecture for Arbitrary Boundary Detectioncited by 12
  9. 2023A Survey of Deep Learning in Sports Applications: Perception, Comprehension, and Decisioncited by 11
  10. 2022A Graph-Based Method for Soccer Action Spotting Using Unsupervised Player Classificationcited by 10
  11. 2022Scene Consistency Representation Learning for Video Scene Segmentationcited by 9
  12. 2025Action Anticipation from SoccerNet Football Video Broadcastscited by 9
  13. 2024Unifying Global and Local Scene Entities Modelling for Precise Action Spottingcited by 8
  14. 2024Towards AI-Powered Video Assistant Referee System (VARS) for Association Footballcited by 8
  15. 2022Going for GOAL: A Resource for Grounded Football Commentariescited by 7
  16. 2023Survey of Action Recognition, Spotting and Spatio-Temporal Localization in Soccer -- Current Trends and Research Perspectivescited by 7
  17. 2024Foul prediction with estimated poses from soccer broadcast videocited by 7
  18. 2024Deep learning for action spotting in association football videoscited by 7
  19. 2025OpenSTARLab: Open Approach for Spatio-Temporal Agent Data Analysis in Soccercited by 7
  20. 2022MMSys'22 Grand Challenge on AI-based Video Production for Soccercited by 6
  21. 2024Action-Agnostic Point-Level Supervision for Temporal Action Detectioncited by 5
  22. 2021A Unified Taxonomy and Multimodal Dataset for Events in Invasion Gamescited by 4
  23. 2022TVNet: Temporal Voting Network for Action Localizationcited by 3
  24. 2025Deep Learning for Sports Video Event Detection: Tasks, Datasets, Methods, and Challengescited by 3
  25. 2021Learning Football Body-Orientation as a Matter of Classificationcited by 2
  26. 2024OSL-ActionSpotting: A Unified Library for Action Spotting in Sports Videoscited by 2
  27. 2025Temporal vs. Spatial: Comparing DINOv3 and V-JEPA2 Feature Representations for Video Action Analysiscited by 2
  28. 2023POTLoc: Pseudo-Label Oriented Transformer for Point-Supervised Temporal Action Localizationcited by 1
  29. 2024Investigating Event-Based Cameras for Video Frame Interpolation in Sportscited by 1
  30. 2025From Broadcast to Minimap: Achieving State-of-the-Art SoccerNet Game State Reconstructioncited by 1
  31. 2025Do We Need Large VLMs for Spotting Soccer Actions?cited by 1
  32. 2026GRAZE: Grounded Refinement and Motion-Aware Zero-Shot Event Localizationcited by 1
  33. 2026Towards Athlete Fatigue Assessment from Association Football Videoscited by 1
  34. 2026MambaKick: Early Penalty Direction Prediction from HAR Embeddingscited by 1
  35. 2026SoccerNet 2026 Challenges Resultscited by 1
  36. 2023Event Detection in Football using Graph Convolutional Networkscited by 0
  37. 2023SoccerKDNet: A Knowledge Distillation Framework for Action Recognition in Soccer Videoscited by 0
  38. 2023ADM-Loc: Actionness Distribution Modeling for Point-supervised Temporal Action Localizationcited by 0

50 most cited of 56 All 56 (JSON)