There is a need to develop accessible 3D motion analysis. The aim of this study was to determine absolute and relative agreement between aiScout®, a smartphone application that infers 3D pose from 2D video, and conventional methods of assessing athletic performance. Sixteen athletes (29.7 ± 9.0 yrs, 1.75 ± 0.10 m, 75.5 ± 15.8 kg) completed six performance tests that were evaluated by aiScout® and conventional methods. 10m sprint times, press-ups, lateral rebound jumps, and 5-10-5 football dribble times showed good to excellent absolute (ICC and CCC > 0.8 threshold) and relative agreement (ICC and Kendall`s W > 0.8 threshold) between measurement methods. Unilateral countermovement jumps only showed moderate agreement (0.6 - 0.8 absolute and relative agreement). This suggests aiScout® can be used to help assess athletic performance. However, questions remain regarding the capability of aiScout® to assess performance of jumping activities.
© Copyright 2024 ISBS Proceedings Archive: Vol. 42: Iss. 1. Alle Rechte vorbehalten.