Journal article
The validity and intra-tester reliability of markerless motion capture to analyse kinematics of the BMX Supercross gate start
Sports Biomechanics, Vol.17(3), pp.383-401
2018
PMID: 29129121
Abstract
The aim of this study was to quantify the validity and intra-tester reliability of a novel method of kinematic measurement. The measurement target was the joint angles of an athlete performing a BMX Supercross (SX) gate start action through the first 1.2 s of movement in situ on a BMX SX ramp using a standard gate start procedure. The method employed GoPro® Hero 4 Silver (GoPro Inc., USA) cameras capturing data at 120 fps 720 p on a 'normal' lens setting. Kinovea 0.8.15 (Kinovea.org, France) was used for analysis. Tracking data was exported and angles computed in Matlab (Mathworks®, USA). The gold standard 3D method for joint angle measurement could not safely be employed in this environment, so a rigid angle was used. Validity was measured to be within 2°. Intra-tester reliability was measured by the same tester performing the analysis twice with an average of 55 days between analyses. Intra-tester reliability was high, with an absolute error <6° and <9 frames (0.075 s) across all angles and time points for key positions, respectively. The methodology is valid within 2° and reliable within 6° for the calculation of joint angles in the first ~1.25 s.
Details
- Title
- The validity and intra-tester reliability of markerless motion capture to analyse kinematics of the BMX Supercross gate start
- Authors
- Josephine Grigg (Author) - Bond UniversityEric Haakonssen (Author) - Australian Institute of SportEvelyne Rathbone (Author) - Bond UniversityRobin Orr (Author) - Bond UniversityJustin W L Keogh (Author) - University of the Sunshine Coast
- Publication details
- Sports Biomechanics, Vol.17(3), pp.383-401
- Publisher
- Routledge
- Date published
- 2018
- DOI
- 10.1080/14763141.2017.1353129
- ISSN
- 1476-3141
- PMID
- 29129121
- Organisation Unit
- University of the Sunshine Coast, Queensland
- Language
- English
- Record Identifier
- 99450446802621
- Output Type
- Journal article
Metrics
1 File views/ downloads
539 Record Views
InCites Highlights
These are selected metrics from InCites Benchmarking & Analytics tool, related to this output
- Collaboration types
- Domestic collaboration
- International collaboration
- Web Of Science research areas
- Engineering, Biomedical
- Sport Sciences