mlsanityData to 7 Oct 2026

Entity-Aware Sequence Transduction for Player-Centric Ball Action Spotting

arXiv 2608.01696 ↗First posted 3 Aug 2026cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Player-centric ball action spotting requires temporally precise event detection together with actor attribution in crowded, partially observed multi-agent sports videos. Existing Denoising Sequence Transduction (DST) baselines treat the player-role dimension as part of a flattened frame-level representation, which weakens the inductive bias for modeling player-specific temporal evolution and inter-player interactions. To address this limitation, we propose Multi-Entity Denoising Sequence Transduction (ME-DST). ME-DST keeps the role-slot dimension throughout encoding. It uses temporal attention to model the history of each role slot, and spatial attention to exchange information across role slots at each frame. This factorized design gives the model a direct structure for separating within-player evolution from inter-player context. We also add learnable role embeddings, tracking-derived tactical features, and fused visual predictions from X3D-L and Swin3D-S. Experiments on the FOOTPASS dataset show that ME-DST reaches a Micro F1 of 0.778. This improves the strongest official TAAD+DST baseline by 10.3 percentage points. Controlled ablations show that preserving the entity axis and encoding role identity are central to this gain. These results suggest that explicit entity modeling is an effective inductive bias for player-centric sports event understanding.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 34 of its 37 references that are in the map. First posted on arXiv 3 Aug 2026.

Builds on

34 of 37 references.

References · 34

  1. 2017Attention Is All You Needcited by 59,932
  2. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,451
  3. 2014Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translationcited by 4,016
  4. 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,127
  5. 2021Video Swin Transformercited by 1,090
  6. 2015NetVLAD: CNN architecture for weakly supervised place recognitioncited by 935
  7. 2020On Layer Normalization in the Transformer Architecturecited by 906
  8. 2016Training Deep Nets with Sublinear Memory Costcited by 825
  9. 2020X3D: Expanding Architectures for Efficient Video Recognitioncited by 635
  10. 2021AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecastingcited by 327
  11. 2020Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Predictioncited by 324
  12. 2018SoccerNet: A Scalable Dataset for Action Spotting in Soccer Videoscited by 148
22 more
  1. 2020SoccerNet-v2: A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videoscited by 129
  2. 2020Actor-Transformers for Group Activity Recognitioncited by 84
  3. 2019A Context-Aware Loss Function for Action Spotting in Soccer Videoscited by 56
  4. 2021GroupFormer: Group Activity Recognition with Clustered Spatial-Temporal Transformercited by 54
  5. 2022Spotting Temporally Precise, Fine-Grained Events in Videocited by 40
  6. 2021Temporally-Aware Feature Pooling for Action Spotting in Soccer Broadcastscited by 30
  7. 2022SoccerNet 2022 Challenges Resultscited by 28
  8. 2022Temporally Precise Action Spotting in Soccer Videos Using Dense Detection Anchorscited by 27
  9. 2024T-DEED: Temporal-Discriminability Enhancer Encoder-Decoder for Precise Event Spotting in Sports Videoscited by 21
  10. 2023SoccerNet 2023 Challenges Resultscited by 20
  11. 2023COMEDIAN: Self-Supervised Learning and Knowledge Distillation for Action Spotting using Transformerscited by 17
  12. 2024ASTRA: An Action Spotting TRAnsformer for Soccer Videoscited by 16
  13. 2022Spatio-Temporal Action Detection Under Large Motioncited by 14
  14. 2025Action Anticipation from SoccerNet Football Video Broadcastscited by 9
  15. 2024SoccerNet 2024 Challenges Resultscited by 8
  16. 2025Game State and Spatio-temporal Action Detection in Soccer using Graph Neural Networks and 3D Convolutional Networkscited by 6
  17. 2025Precise Event Spotting in Sports Videos: Solving Long-Range Dependency and Class Imbalancecited by 6
  18. 2025Beyond Pixels: Leveraging the Language of Soccer to Improve Spatio-Temporal Action Detection in Broadcast Videoscited by 5
  19. 2025FOOTPASS: A Multi-Modal Multi-Agent Tactical Context Dataset for Play-by-Play Action Spotting in Soccer Broadcast Videoscited by 5
  20. 2023A Boosted Model Ensembling Approach to Ball Action Spotting in Videos: The Runner-Up Solution to CVPR'23 SoccerNet Challengecited by 4
  21. 2025SoccerNet 2025 Challenges Resultscited by 4
  22. 2026Decoding Defensive Coverage Responsibilities in American Football Using Factorized Attention Based Transformer Modelscited by 1

Built on by

Not cited yet.