mlsanityData to 14 Oct 2026

AthletePose3D: A Benchmark Dataset for 3D Human Pose Estimation and Kinematic Validation in Athletic Movements

arXiv 2503.07499 ↗First posted 10 Mar 2025cs.CVcited by 14

Not on the map. Why?

Abstract (arXiv)

Human pose estimation is a critical task in computer vision and sports biomechanics, with applications spanning sports science, rehabilitation, and biomechanical research. While significant progress has been made in monocular 3D pose estimation, current datasets often fail to capture the complex, high-acceleration movements typical of competitive sports. In this work, we introduce AthletePose3D, a novel dataset designed to address this gap. AthletePose3D includes 12 types of sports motions across various disciplines, with approximately 1.3 million frames and 165 thousand individual postures, specifically capturing high-speed, high-acceleration athletic movements. We evaluate state-of-the-art (SOTA) monocular 2D and 3D pose estimation models on the dataset, revealing that models trained on conventional datasets perform poorly on athletic motions. However, fine-tuning these models on AthletePose3D notably reduces the SOTA model mean per joint position error (MPJPE) from 214mm to 65mm-a reduction of over 69%. We also validate the kinematic accuracy of monocular pose estimations through waveform analysis, highlighting strong correlations in joint angle estimations but limitations in velocity estimation. Our work provides a comprehensive evaluation of monocular pose estimation models in the context of sports, contributing valuable insights for advancing monocular pose estimation techniques in high-performance sports environments. The dataset, code, and model checkpoints are available at: https://github.com/calvinyeungck/AthletePose3D

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Builds on 60 of its 89 references that are in the map. First posted on arXiv 10 Mar 2025.

Builds on

60 of 89 references.

References · 60

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  5. 2022A ConvNet for the 2020scited by 3,579
  6. 2019Deep High-Resolution Representation Learning for Human Pose Estimationcited by 1,339
  7. 2018OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,162
  8. 2016Realtime Multi-Person 2D Pose Estimation using Part Affinity Fieldscited by 1,064
  9. 2019MediaPipe: A Framework for Building Perception Pipelinescited by 583
  10. 2018Simple Baselines for Human Pose Estimation and Trackingcited by 502
  11. 2016Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervisioncited by 494
  12. 2021AI Choreographer: Music Conditioned 3D Dance Generation with AIST++cited by 485
38 more
  1. 2022ViTPose: Simple Vision Transformer Baselines for Human Pose Estimationcited by 456
  2. 20183D human pose estimation in video with temporal convolutions and semi-supervised trainingcited by 403
  3. 2017A simple yet effective baseline for 3d human pose estimationcited by 401
  4. 2023Effective Whole-body Pose Estimation with Two-stages Distillationcited by 311
  5. 2013DeepPose: Human Pose Estimation via Deep Neural Networkscited by 305
  6. 20213D Human Pose Estimation with Spatial and Temporal Transformerscited by 272
  7. 2019HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimationcited by 247
  8. 2022MotionBERT: A Unified Perspective on Learning Human Motion Representationscited by 193
  9. 2022Focal Modulation Networkscited by 186
  10. 2022UniFormer: Unified Transformer for Efficient Spatiotemporal Representation Learningcited by 184
  11. 2023RTMPose: Real-Time Multi-Person Pose Estimation based on MMPosecited by 175
  12. 2019Learnable Triangulation of Human Posecited by 161
  13. 2022MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Videocited by 159
  14. 2021MHFormer: Multi-Hypothesis Transformer for 3D Human Pose Estimationcited by 151
  15. 2021HRFormer: High-Resolution Transformer for Dense Predictioncited by 149
  16. 2022More ConvNets in the 2020s: Scaling up Kernels Beyond 51x51 using Sparsitycited by 142
  17. 2014Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimationcited by 129
  18. 2020TransPose: Keypoint Localization via Transformercited by 126
  19. 2021Bottom-Up Human Pose Estimation Via Disentangled Keypoint Regressioncited by 112
  20. 2018Pose2Seg: Detection Free Human Instance Segmentationcited by 107
  21. 2021TokenPose: Learning Keypoint Tokens for Human Pose Estimationcited by 107
  22. 20213D Human Action Representation Learning via Cross-View Consistency Pursuitcited by 106
  23. 2022HorNet: Efficient High-Order Spatial Interactions with Recursive Gated Convolutionscited by 103
  24. 2021Pose Recognition with Cascade Transformerscited by 99
  25. 2021Exploiting Temporal Contexts with Strided Transformer for 3D Human Pose Estimationcited by 94
  26. 2023PoseFormerV2: Exploring Frequency Domain for Efficient and Robust 3D Human Pose Estimationcited by 92
  27. 2020Motion Guided 3D Pose Estimation from Videoscited by 85
  28. 2023Diffusion-Based 3D Human Pose Estimation with Multi-Hypothesis Aggregationcited by 82
  29. 2022P-STMO: Pre-Trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimationcited by 80
  30. 2023MotionAGFormer: Enhancing 3D Human Pose Estimation with a Transformer-GCNFormer Networkcited by 64
  31. 2022MogaNet: Multi-order Gated Aggregation Networkcited by 59
  32. 2022GFPose: Learning 3D Human Pose Prior with Gradient Fieldscited by 57
  33. 20203D Human Pose Estimation using Spatio-Temporal Networks with Explicit Occlusion Trainingcited by 54
  34. 2020TransMoMo: Invariance-Driven Unsupervised Video Motion Retargetingcited by 36
  35. 2024KTPFormer: Kinematics and Trajectory Prior Knowledge-Enhanced Transformer for 3D Human Pose Estimationcited by 34
  36. 2022Uplift and Upsample: Efficient 3D Human Pose Estimation with Uplifting Transformerscited by 33
  37. 2023SportsPose -- A Dynamic 3D sports pose datasetcited by 26
  38. 2021Contrast-reconstruction Representation Learning for Self-supervised Skeleton-based Action Recognitioncited by 19

50 most cited of 60 All 60 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 9.

14 papers.

Sort:

  1. 2025Gen4D: Synthesizing Humans and Scenes in the Wildcited by 3
  2. 2026Breaking Coordinate Overfitting: Geometry-Aware WiFi Sensing for Cross-Layout 3D Pose Estimationcited by 2
  3. 2025Multi-person Physics-based Pose Estimation for Combat Sportscited by 1
  4. 2026CalTennis: Large Multi-View Tennis Video Dataset and Benchmark of Monocular-to-3D Pose Estimationcited by 1
  5. 2025KASportsFormer: Kinematic Anatomy Enhanced Transformer for 3D Human Pose Estimation on Short Sports Scene Videocited by 0
  6. 2025PRGCN: A Graph Memory Network for Cross-Sequence Pattern Reuse in 3D Human Pose Estimationcited by 0
  7. 2025A Topology-Aware Graph Convolutional Network for Human Pose Similarity and Action Quality Assessmentcited by 0
  8. 2026A Study on Real-time Object Detection using Deep Learningcited by 0
  9. 2026CPU Optimization of a Monocular 3D Biomechanics Pipeline for Low-Resource Deploymentcited by 0
  10. 2026Mesh-Aware Epipolar Matching for Multi-View Multi-Person 3D Pose Estimation in Basketballcited by 0
  11. 2026TaskNPoint: How to Teach Your Humanoid to Hit a Backhand in Minutescited by 0
  12. 2026Markerless Pose Estimation for Resistance Training Technique Assessmentcited by 0
2 more
  1. 2026MuyBridge: Mobile Human Center-of-Mass Estimation from Monocular Video via Sparse Fusioncited by 0
  2. 2026Field Converter: Geometry-Initialized Temporal Residual Refinement for World-Grounded Player Pose Estimation from Soccer Broadcastscited by 0