ASSESSMENT OF KINEMATIC CMJ DATA USING A DEEP LEARNING ALGORITHM-BASED MARKERLESS MOTION CAPTURE SYSTEM

Authors

Gerda Strutzenberger, Department of Sport and Exercise Science, University of Salzburg, Austria; Sports Medical Research Group, Department of Orthopaedics, Balgrist University Hospital, University of Zurich, Switzerland2Follow
Robert Kanko, Department of Mechanical & Materials Engineering, Queen's University, Kingston, Canada
Scott Selbie, Theia Markerless Inc, Kingston, Canada
Hermann Schwameder, Theia Markerless Inc, Kingston, Canada
Kevin Deluzio, Department of Mechanical & Materials Engineering, Queen's University, Kingston, Canada

Category

Methods / Statistics

Document Type

Paper

Abstract

The purpose of this study was to compare the performance of a video-based markerless motion capture system to a conventional marker-based approach during a counter movement jump (CMJ). Twenty-three healthy participants performed CMJ while data was collected simultaneously via a marker-based (Oqus) and a 2D video-based motion capture system (Miqus, both: Qualisys). The video data was further processed to 3D-data using Theia3D (Theia Markerless Inc.). Excellent agreement between systems with ICCs >0.99 exists for jump height (mean average error of -0.27 cm) and sagittal ankle and knee plane angles (RMSD < 5°). The hip joint showed an average RMSD of 21° with a strong correlation of 0.80. As such the markerless system is capable of detecting jump height, sagittal ankle and knee joint angles and 3D joint positions of a CMJ to a high accuracy

Recommended Citation

Strutzenberger, Gerda; Kanko, Robert; Selbie, Scott; Schwameder, Hermann; and Deluzio, Kevin (2021) "ASSESSMENT OF KINEMATIC CMJ DATA USING A DEEP LEARNING ALGORITHM-BASED MARKERLESS MOTION CAPTURE SYSTEM," ISBS Proceedings Archive: Vol. 39: Iss. 1, Article 61.
Available at: https://commons.nmu.edu/isbs/vol39/iss1/61