A Study of Starting Correction in Sports Based in Human Posture Estimation and Attention Mechanism
Abstract
The standardisation and accuracy of the starting movement will affect the athletes' performance and the results of the game, and the differences in different sports, athletes' physical qualities and coaches' teaching methods make it difficult for the traditional correction of the starting movement to reflect a better application effect. Therefore, the study is based on the Openpose human posture estimation and recognition algorithm, which performs frame-posture relationship correlation analysis, continuity movement design and joint bone point repair to improve the accuracy of recognition. At the same time, taking into account the differences in the content of different assignments of posture data information, the study introduces a lightweight attention mechanism that focuses on temporal patterns on the basis of the recognition algorithm. The experimental results show that the recognition accuracy of the improved algorithm for the starting posture is above 0.85, and its recognition accuracy for the three phases of the starting action is above 93% compared with the Dynamic Time Warping (DTW) model and the High Resolution Human Keypoint Detection Network (HRNet-w48), with an F1 value of 0.6857, which is better than that of the DTW model (0.6857), and the F1 value of 0.6857. better than the DTW model (0.6214) and HRNet-w48 model (0.6321). The detection and recognition time of the improved algorithm for the starting movement is less than 0.1s, and it can better provide athletes with movement correction suggestions and formulate personalised training plans.
Full Text:
PDFReferences
Ma N, Wu Z, Cheung Y, Guo Y, Gao Y, Li J, et al. A survey of human action recognition and posture prediction. Tsinghua Sci Technol. 2022; 27(6): 973-1001.
Dorofieieva O, Yarymbash K, Skrypchenko I, Joksimovi M, and Mytsak A. Complex assessment of athletes' operative status and its correction during competitions, based on the body impedance analysis. Pedagogy Phys Cult Sports. 2020; 25(2):66-73.
Li Y, Guo T, Liu X, Luo W, and Xie W. Action status based novel relative feature representations for interaction recognition. Acta Electronica Sinica. 2022; 31(1):168-180.
Zhang J, Ye G, Tu Z, Qin Y, Qin Q, Zhang J, et al. A spatial attentive and temporal dilated (SATD) GCN for skeleton-based action recognition. Journal of Intelligent Technology. 2022; 7(1):46-55.
Tani Y, Tasaki R, Terashima K. Walking tracking system based on estimation of human posture in omni-directional mobile walker. J. Meteorol. Soc. Japan. 2019; 37(2):161-167.
Le V. Deep learning-based for human segmentation and tracking, 3D human pose estimation and action recognition on monocular video of MADS dataset. Multimed Tools Appl. 2023; 82(14): 20771-20818.
Xu J, Liu W, Xing W, and Wei X. MSPENet: multi-scale adaptive fusion and position enhancement network for human pose estimation. VC. 2022; 39(5):2005-2019.
Kumar P, Chauhan S, Awasthi L. Human pose estimation using deep learning: review, methodologies, progress and future research directions. Int J Multimed Inf R. 2022; 11(4):489-521.
Huang L, Liu G. Functional motion detection based on artificial intelligence. J Supercomput. 2022; 78(3): 4290-4329.
Wang X, Wu Y, Zhu L, and Yang Y. Symbiotic attention with privileged information for egocentric action recognition. Proceedings of the AAAI Conference on Artificial Intelligence. 2020; 34(7):12249-12256.
Zhou K, Wu T, Wang C, Wang J, and Li C. Skeleton based abnormal behavior recognition using spatio-temporal convolution and attention-based LSTM. Procedia Computer Science. 2020; 174(8):424-432.
Li Y, Yuan J, Liu H. Human skeleton-based action recognition algorithm based on spatiotemporal attention graph convolutional network model. Journal of Computer Applications. 2021; 41(7):1915-1921.
Li X, Xie M, Zhang Y, Ding G, and Tong W. Dual attention convolutional network for action recognition. IET Image Processing. 2020; 14(6):1059-1065.
Zhang Y. Fused behaviour recognition model based on attention mechanism. Visual Computing in Industrial and Medical Arts. 2020; 3(1):71-80.
Wu J, Xue Y, Meng X, Wan X. Research on behaviour recognition algorithm based on SE-I3D-GRU networkI. High Technol. Lett. 2021; 27(2):163-172.
Wang Z. Spatial-Temporal feature-based sports video classification. IJACI. 2021; 12(4):79-97.
Wu S. Image recognition of standard actions in sports videos based on feature fusion. Traitement du Signal. 2021; 38(6):1801- 1807.
Agethen S, Hsu W. Deep multi-kernel convolutional lstm networks and an attention-based mechanism for videos. IEEE. 2020; 22(3):819-829.
Liu J, Che Y. Action recognition for sports video analysis using part-attention spatio-temporal graph convolutional network. J Electron Imaging. 2021; 30(3):33017.1-33017.16.
Fang L, Sun M. Motion recognition technology of badminton players in sports video images. FGCS. 2021; 124(9):381- 389.
Chen Z. Research on internet security situation awareness prediction technology based on improved RBF neural network algorithm. JCCE. 2022; 1(3): 103-108.
Cai X. WSN-Driven Posture recognition and correction towards basketball exercise. International journal of Information System Modeling and Design, 2022.
Fu D. Research on intelligent recognition technology of gymnastics posture based on KNN fusion DTW algorithm based on sensor technology. International Journal of Wireless and Mobile Computing: IJWMC, 2023.
DOI: https://doi.org/10.31449/inf.v48i23.6759
This work is licensed under a Creative Commons Attribution 3.0 License.