Article: The application of VR-based fine motion capture algorithm in college aerobics training Journal: International Journal of Computational Systems Engineering (IJCSYSE) 2025 Vol.9 No.6 pp.1 - 10 Abstract: In response to the problems of noise and incompleteness in the motion capture of VR technology in college aerobics training, this study first built a fine motion capture model based on an improved iterative nearest point algorithm, and then constructed an action recognition model based on an improved spatiotemporal graph convolutional neural network. The outcomes denote that the Top-1 of the action capture model is 36.5%, while the Top-5 is 59.4%. The Top-1 and Top-5 accuracy peaks of the action recognition model are 90.1% and 99.0%, respectively. The classification accuracy on the two datasets is 0.914 and 0.983, respectively. The standardisation level of the experimental group is 9.4 points higher than that of the control group. In summary, the model constructed through research has good application effects in fine motion capture and recognition, which helps to improve the teaching effect of efficient aerobics. Inderscience Publishers - linking academia, business and industry through research

Title: The application of VR-based fine motion capture algorithm in college aerobics training

Authors: Hui Wang

Addresses: School of Physical Education, Yan'an University, Yan'an 716000, China

Abstract: In response to the problems of noise and incompleteness in the motion capture of VR technology in college aerobics training, this study first built a fine motion capture model based on an improved iterative nearest point algorithm, and then constructed an action recognition model based on an improved spatiotemporal graph convolutional neural network. The outcomes denote that the Top-1 of the action capture model is 36.5%, while the Top-5 is 59.4%. The Top-1 and Top-5 accuracy peaks of the action recognition model are 90.1% and 99.0%, respectively. The classification accuracy on the two datasets is 0.914 and 0.983, respectively. The standardisation level of the experimental group is 9.4 points higher than that of the control group. In summary, the model constructed through research has good application effects in fine motion capture and recognition, which helps to improve the teaching effect of efficient aerobics.

Keywords: action capture algorithm; virtual reality; VR; aerobics training; iteration closest point.

DOI: 10.1504/IJCSYSE.2025.145446

International Journal of Computational Systems Engineering, 2025 Vol.9 No.6, pp.1 - 10

Received: 13 Nov 2023
Accepted: 08 Feb 2024
Published online: 01 Apr 2025 *

Title: The application of VR-based fine motion capture algorithm in college aerobics training

Keep up-to-date