mlsanityData to 14 Oct 2026

AutoSoccerPose: Automated 3D posture Analysis of Soccer Shot Movements

arXiv 2405.12070 ↗First posted 20 May 2024cs.CV cs.AIcited by 15

Not on the map. Why?

Abstract (arXiv)

Image understanding is a foundational task in computer vision, with recent applications emerging in soccer posture analysis. However, existing publicly available datasets lack comprehensive information, notably in the form of posture sequences and 2D pose annotations. Moreover, current analysis models often rely on interpretable linear models (e.g., PCA and regression), limiting their capacity to capture non-linear spatiotemporal relationships in complex and diverse scenarios. To address these gaps, we introduce the 3D Shot Posture (3DSP) dataset in soccer broadcast videos, which represents the most extensive sports image dataset with 2D pose annotations to our knowledge. Additionally, we present the 3DSP-GRAE (Graph Recurrent AutoEncoder) model, a non-linear approach for embedding pose sequences. Furthermore, we propose AutoSoccerPose, a pipeline aimed at semi-automating 2D and 3D pose estimation and posture analysis. While achieving full automation proved challenging, we provide a foundational baseline, extending its utility beyond the scope of annotated data. We validate AutoSoccerPose on SoccerNet and 3DSP datasets, and present posture analysis results based on 3DSP. The dataset, code, and models are available at: https://github.com/calvinyeungck/3D-Shot-Posture-Dataset.

Key facts. Cited by 15 papers in the map (data through 14 Oct 2026). Builds on 40 of its 59 references that are in the map. First posted on arXiv 20 May 2024.

Builds on

40 of 59 references.

References · 40

  1. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  3. 2015You Only Look Once: Unified, Real-Time Object Detectioncited by 4,852
  4. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  5. 2021YOLOX: Exceeding YOLO Series in 2021cited by 1,071
  6. 2016Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,064
  7. 2017Simple Online and Realtime Tracking with a Deep Association Metriccited by 910
  8. 2021ByteTrack: Multi-Object Tracking by Associating Every Detection Boxcited by 823
  9. 2020HOTA: A Higher Order Metric for Evaluating Multi-Object Trackingcited by 552
  10. 2023DETRs Beat YOLOs on Real-time Object Detectioncited by 533
  11. 2021TrackFormer: Multi-Object Tracking with Transformerscited by 518
  12. 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 494
28 more
  1. 2015Unsupervised Learning of Video Representations using LSTMscited by 435
  2. 2022Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Trackingcited by 346
  3. 2022BoT-SORT: Robust Associations Multi-Pedestrian Trackingcited by 320
  4. 2023Effective Whole-body Pose Estimation with Two-stages Distillationcited by 311
  5. 2013DeepPose: Human Pose Estimation via Deep Neural Networkscited by 305
  6. 2019HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimationcited by 247
  7. 2020Whole-Body Human Pose Estimation in the Wildcited by 178
  8. 2023RTMPose: Real-Time Multi-Person Pose Estimation based on MMPosecited by 175
  9. 2019Learnable Triangulation of Human Posecited by 161
  10. 2018SoccerNet: A Scalable Dataset for Action Spotting in Soccer Videoscited by 150
  11. 2022MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectorscited by 148
  12. 2020SoccerNet-v2: A Dataset and Benchmarks for Holistic Understanding of Broadcast Soccer Videoscited by 131
  13. 2014Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimationcited by 129
  14. 2017Towards 3D Human Pose Estimation in the Wild: a Weakly-supervised Approachcited by 115
  15. 2021Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regressioncited by 112
  16. 2018Pose2Seg: Detection Free Human Instance Segmentationcited by 107
  17. 2020Motion Guided 3D Pose Estimation from Videoscited by 85
  18. 2023MotionAGFormer: Enhancing 3D Human Pose Estimation with a Transformer-GCNFormer Networkcited by 64
  19. 2023VARS: Video Assistant Referee System for Automated Soccer Decision Making from Multiple Viewscited by 38
  20. 2023SportsPose -- A Dynamic 3D sports pose datasetcited by 26
  21. 2023SoccerNet 2023 Challenges Resultscited by 20
  22. 2022Learnable human mesh triangulation for 3D human pose and shape estimationcited by 16
  23. 2023Transformer-Based Neural Marked Spatio Temporal Point Process Model for Football Match Events Analysiscited by 10
  24. 2023A Strategic Framework for Optimal Decisions in Football 1-vs-1 Shot-Taking Situations: An Integrated Approach of Machine Learning, Theory-Based Modeling, and Game Theorycited by 7
  25. 2023Fine-Grained Sports, Yoga, and Dance Postures Recognition: A Benchmark Analysiscited by 7
  26. 2024Foul prediction with estimated poses from soccer broadcast videocited by 7
  27. 2023Automatic Edge Error Judgment in Figure Skating Using 3D Pose Estimation from a Monocular Camera and IMUscited by 6
  28. 2022Learning from the Pros: Extracting Professional Goalkeeper Technique from Broadcast Footagecited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 1; 2025: 11; 2026: 3.

15 papers.

Sort:

  1. 2025AthletePose3D: A Benchmark Dataset for 3D Human Pose Estimation and Kinematic Validation in Athletic Movementscited by 14
  2. 20243D Pose-Based Temporal Action Segmentation for Figure Skating: A Fine-Grained and Jump Procedure-Aware Annotation Approachcited by 7
  3. 2025OpenSTARLab: Open Approach for Spatio-Temporal Agent Data Analysis in Soccercited by 7
  4. 2025AthleticsPose: Authentic Sports Motion Dataset on Athletic Field and Evaluation of Monocular 3D Pose Estimation Abilitycited by 6
  5. 2025TrackID3x3: A Dataset and Algorithm for Multi-Player Tracking with Identification and Pose Estimation in 3x3 Basketball Full-court Videoscited by 4
  6. 2025SoccerNet-v3D: Leveraging Sports Broadcast Replays for 3D Scene Understandingcited by 4
  7. 2025Benchmarking 3D Human Pose Estimation Models under Occlusionscited by 3
  8. 2025BGM2Pose: Active 3D Human Pose Estimation with Non-Stationary Soundscited by 1
  9. 2026ExpertEdit: Learning Skill-Aware Motion Editing from Expert Videoscited by 1
  10. 2025Polar Coordinate-Based 2D Pose Prior with Neural Distance Fieldcited by 0
  11. 2025KASportsFormer: Kinematic Anatomy Enhanced Transformer for 3D Human Pose Estimation on Short Sports Scene Videocited by 0
  12. 2025VIFSS: View-Invariant and Figure Skating-Specific Pose Representation Learning for Temporal Action Segmentationcited by 0
3 more
  1. 2025Avatar4D: Synthesizing Domain-Specific 4D Humans for Real-World Pose Estimationcited by 0
  2. 2026Mesh-Aware Epipolar Matching for Multi-View Multi-Person 3D Pose Estimation in Basketballcited by 0
  3. 2026Field Converter: Geometry-Initialized Temporal Residual Refinement for World-Grounded Player Pose Estimation from Soccer Broadcastscited by 0