Video skeletonization and AI: a novel approach for analyzing kinematics in table tennis

This study introduces a simple yet effective system to measure table tennis player movements and relate them to the moment the ball hits the racket. The system accurately measures shoulder, elbow, hip and knee joint angles without hindering player movements, using only two GoPro Hero 10 cameras and the Mediapipe framework. Ten male participants were evaluated using synchronized cameras and a custom-trained YOLOv8 algorithm to pinpoint the moment of impact between the racket and the ball. This approach, avoiding participant-worn equipment, offers a streamlined solution for assessing table tennis movements, promising potential applications in sports analysis.
© Copyright 2024 ISBS Proceedings Archive: Vol. 42: Iss. 1. All rights reserved.

Bibliographic Details
Subjects:
Notations:technical and natural sciences sport games
Tagging:künstliche Intelligenz markerless
Published in:ISBS Proceedings Archive: Vol. 42: Iss. 1
Language:English
Published: 2024
Online Access:https://commons.nmu.edu/isbs/vol42/iss1/219/
Volume:42
Issue:1
Pages:219
Document types:article
Level:advanced