mlsanityData to 7 Oct 2026

Temporally Precise Action Spotting in Soccer Videos Using Dense Detection Anchors

arXiv 2205.10450 ↗First posted 20 May 2022cs.CVcited by 27

Not on the map. Why?

Abstract (arXiv)

We present a model for temporally precise action spotting in videos, which uses a dense set of detection anchors, predicting a detection confidence and corresponding fine-grained temporal displacement for each anchor. We experiment with two trunk architectures, both of which are able to incorporate large temporal contexts while preserving the smaller-scale features required for precise localization: a one-dimensional version of a u-net, and a Transformer encoder (TE). We also suggest best practices for training models of this kind, by applying Sharpness-Aware Minimization (SAM) and mixup data augmentation. We achieve a new state-of-the-art on SoccerNet-v2, the largest soccer video dataset of its kind, with marked improvements in temporal localization. Additionally, our ablations show: the importance of predicting the temporal displacements; the trade-offs between the u-net and TE trunks; and the benefits of training with SAM and mixup.

Key facts. Cited by 27 papers in the map (data through 7 Oct 2026). Builds on 18 of its 22 references that are in the map. First posted on arXiv 20 May 2022.

Builds on

18 of 22 references.

References · 18

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  4. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,155
  5. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  6. 2020Sharpness-Aware Minimization for Efficiently Improving Generalizationcited by 1,404
  7. 2020Long Range Arena: A Benchmark for Efficient Transformerscited by 696
  8. 2021When Vision Transformers Outperform ResNets without Pre-training or Strong Data Augmentationscited by 259
  9. 2018SoccerNet: A Scalable Dataset for Action Spotting in Soccer Videoscited by 148
  10. 2020SoccerNet-v2: A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videoscited by 129
  11. 2019A Context-Aware Loss Function for Action Spotting in Soccer Videoscited by 56
  12. 2021Camera Calibration and Player Localization in SoccerNet-v2 and Investigation of their Representations for Action Spottingcited by 43
6 more
  1. 2021Temporally-Aware Feature Pooling for Action Spotting in Soccer Broadcastscited by 30
  2. 2021Feature Combination Meets Attention: Baidu Soccer Embeddings and Transformer based Temporal Detectioncited by 27
  3. 2020Improved Soccer Action Spotting using both Audio and Video Streamscited by 21
  4. 2021RMS-Net: Regression and Masking for Soccer Event Spottingcited by 19
  5. 2020Event detection in coarsely annotated sports videos via parallel multi receptive field 1D convolutionscited by 18
  6. 2022Faster-TAD: Towards Temporal Action Detection with Proposal Generation and Classification in a Unified Networkcited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 7; 2024: 10; 2025: 3; 2026: 4.

27 papers.

Sort:

By area:

  1. 2023SoccerNet-Caption: Dense Video Captioning for Soccer Broadcasts Commentariescited by 57
  2. 2022Spotting Temporally Precise, Fine-Grained Events in Videocited by 40
  3. 2023VARS: Video Assistant Referee System for Automated Soccer Decision Making from Multiple Viewscited by 38
  4. 2024SoccerNet Game State Reconstruction: End-to-End Athlete Tracking and Identification on a Minimapcited by 34
  5. 2022SoccerNet 2022 Challenges Resultscited by 28
  6. 2024X-VARS: Introducing Explainability in Football Refereeing with Multi-Modal Large Language Modelcited by 22
  7. 2024T-DEED: Temporal-Discriminability Enhancer Encoder-Decoder for Precise Event Spotting in Sports Videoscited by 21
  8. 2023SoccerNet 2023 Challenges Resultscited by 20
  9. 2023COMEDIAN: Self-Supervised Learning and Knowledge Distillation for Action Spotting using Transformerscited by 17
  10. 2024ASTRA: An Action Spotting TRAnsformer for Soccer Videoscited by 16
  11. 2022Action Spotting using Dense Detection Anchors Revisited: Submission to the SoccerNet Challenge 2022cited by 14
  12. 2023Towards Active Learning for Action Spotting in Association Football Videoscited by 14
15 more
  1. 2023A Survey of Deep Learning in Sports Applications: Perception, Comprehension, and Decisioncited by 11
  2. 2025Action Anticipation from SoccerNet Football Video Broadcastscited by 9
  3. 2024Unifying Global and Local Scene Entities Modelling for Precise Action Spottingcited by 8
  4. 2024Towards AI-Powered Video Assistant Referee System (VARS) for Association Footballcited by 8
  5. 2024SoccerNet 2024 Challenges Resultscited by 8
  6. 2023Survey of Action Recognition, Spotting and Spatio-Temporal Localization in Soccer -- Current Trends and Research Perspectivescited by 7
  7. 2024Deep learning for action spotting in association football videoscited by 7
  8. 2025Precise Event Spotting in Sports Videos: Solving Long-Range Dependency and Class Imbalancecited by 6
  9. 2025Deep Learning for Sports Video Event Detection: Tasks, Datasets, Methods, and Challengescited by 3
  10. 2024OSL-ActionSpotting: A Unified Library for Action Spotting in Sports Videoscited by 2
  11. 2026AdaSpot: Spend Resolution Where It Matters for Precise Event Spottingcited by 2
  12. 2024Investigating Event-Based Cameras for Video Frame Interpolation in Sportscited by 1
  13. 2026Towards Athlete Fatigue Assessment from Association Football Videoscited by 1
  14. 2026SoccerNet 2026 Challenges Resultscited by 1
  15. 2026Entity-Aware Sequence Transduction for Player-Centric Ball Action Spottingcited by 0