Application of deep learning neural networks for detecting spatial key points of a human when performing sports exercises

Abstract


This paper discusses the use of neural networks to detect spatial key points of a person when performing sports exercises. Key point detection technology allows you to track the movements of athletes in real time, conduct an in-depth analysis of their technique and automate the execution of exercises. This helps coaches identify weaknesses and improve athletes' skills. The main attention is paid to methods of 2D and 3D detection of key points, their application in sports and efficiency analysis. The results of 3D detection of key points for an athlete performing an exercise are presented.

Full Text

5

About the authors

A. D Teryohin

Perm National Research Polytechnic University

S. A Fedoseev

Perm National Research Polytechnic University

V. Yu Stolbov

Perm National Research Polytechnic University

References

  1. Self-Supervised Learning for Human Pose Estimation in Sports / K. Ludwig, S. Scherer, M. Einfalt, R. Lienhart // International Conference on Multimedia & Expo Workshops. – 2021. doi: 10.1109/ICMEW53276.2021.9456000
  2. SportsPose – A Dynamic 3D sports pose dataset / C. Ingwersen, C. Moller Mikkelstrup, J. Jensen, M. Rieger Hannemose, A. Dahl // Conference on Computer Vision and Pattern Recognition Workshops. – 2023. doi: 10.1109/CVPRW59228.2023.00550
  3. Athlete Pose Detection Using OpenCV in Deep Learning [Электронный ресурс]. – URL: https://medium.com/swlh/athlete-pose-detection-3d1b93f2d82e (дата обращения: 20.05.2024).
  4. Realtime Multi-person 2D Pose Estimation Using Part Affinity Fields / Z. Cao, T. Simon, S. Wei, Y. Sheikh // Conference on Computer Vision and Pattern Recognition. – 2017. doi: 10.1109/CVPR.2017.143
  5. Deep High-Resolution Representation Learning for Human Pose Estimation / K. Sun, B. Xiao, D. Liu, J. Wang // Conference on Computer Vision and Pattern Recognition. – 2019. doi: 10.1109/CVPR.2019.00584
  6. Microsoft COCO: Common Objects in Context [Электронный ресурс] / Tsung-Yi Lin, Michael Maire, Serge Belongie, Lubomir Bourdev, Ross Girshick, James Hays, Pietro Perona, Deva Ramanan, C. Lawrence Zitnick, Piotr Dollár. – URL: https://arxiv.org/abs/1405.0312 (дата обращения: 05.05.2024).
  7. D Human Pose Estimation: New Benchmark and State of the Art Analysis / M. An-driluka, L. Pishchulin, P. Gehler, S. Bernt // Conference on Computer Vision and Pattern Recognition. – 2014. doi: 10.1109/CVPR.2014.471
  8. Серов, С.С. Метод распознавания игровых поз спортсменов на видео на основе траекторий перемещения и скелетных представлений / С.С. Серов, Д.Р. Пойчанов, Г.И. Афанасьев // Физико-техническая информатика: сборник трудов конференции. – Пущино, 2022. – С. 170–172.
  9. Action Recognition and Sports Evaluation of Running Pose Based on Pose Estimation / Y. Yang, Y. Zeng, L. Yang, Y. Lu, X. Lee, Y. Enomoto // International Journal of Human Movement and Sports Sciences. – 2024. doi: 10.13189/saj.2024.120118
  10. Pose estimation and motion analysis of ski jumpers based on ECA-HRNet / W. Bao, T. Niu, N. Wang, X. Yang // Scientific Reports. – 2023. doi: 10.1038/s41598-023-32893-x
  11. Терехин, А.Д. Система оценивания спортивных упражнений по нейросетевому анализу видеоряда / А.Д. Терехин, О.Р. Ильялов, А.В. Степанов // Прикладная математика и вопросы управления = Applied Mathematics and Control Sciences. – 2022. – № 1. – С. 75–86. doi: 10.15593/2499-9873/2022.1.04
  12. Osokin, D. Real-time 2D Multi-Person Pose Estimation on CPU: Lightweight OpenPose [Электронный ресурс] / D. Osokin. – URL: https://arxiv.org/abs/1811.12004 (дата обращения: 29.04.2024).
  13. Mehraban, S. MotionAGFormer: Enhancing 3D Human Pose Estimation with a Transformer-GCNFormer Network / S. Mehraban, V. Adeli, B. Taati // Winter Conference on Applications of Computer Vision. – 2024. doi: 10.1109/WACV57701.2024.00677
  14. Chun, S. Representation learning of vertex heatmaps for 3D human mesh reconstruction from multi-view images [Электронный ресурс] / S. Chun, S. Park, J. Yong Chang. – URL: https://arxiv.org/pdf/2306.16615v1.pdf (дата обращения: 29.04.2024).
  15. D human pose estimation with spatial and temporal transformers / C. Zheng, S. Zhu, M. Mendieta, T. Yang, C. Chen, Z. Ding // International Conference on Computer Vision. – 2021. doi: 10.1109/ICCV48922.2021.01145
  16. Human 3.6M: Large Scale Datasets and Predictive Methods for 3D Human Sensing in Natural Environments / C. Ionescu, D. Papava, V. Olaru, C. Sminchisescu // Transactions on Pattern Analysis and Machine Intelligence. – 2014. doi: 10.1109/TPAMI.2013.248
  17. Monocular 3D Human Pose Estimation in the Wild Using Improved CNN Supervision / D. Mehta, H. Rhodin, D. Casas, P. Fua, O. Sotnychenko, W. Xu, C. Theobalt // International Conference on 3D Vision (3DV). – 2017. doi: 10.1109/3DV.2017.00064
  18. Vision-based Pose Estimation for Augmented Reality: A Comparison Study [Электронный ресурс] / H. Belghit, A. Bellarbi, N. Zenati-Henda, S. Otmane. – URL: https://arxiv.org/pdf/1806.09316 (дата обращения: 29.04.2024).
  19. Baumgartner, T. Monocular 3D Human Pose Estimation for Sports Broadcasts using Partial Sports Field Registration / T. Baumgartner, S. Klatt // Conference on Computer Vision and Pattern Recognition Workshops. – 2023. doi: 10.1109/CVPRW59228.2023.00539

Statistics

Views

Abstract - 5

PDF (Russian) - 1

Refbacks

  • There are currently no refbacks.

This website uses cookies

You consent to our cookies if you continue to use our website.

About Cookies