Related Experiment Video
Updated: Mar 17, 2026
04:06
Author Spotlight: Enhancing Remote Rehabilitation with Virtual Reality and Electromyography
Published on: January 12, 2024
1.1K
Development of a Kinect Software Tool to Classify Movements during Active Video Gaming
Michael Rosenberg1, Ashleigh L Thornton1, Brendan S Lay1
1School of Sport Science, Exercise and Health, University of Western Australia, M408 35 Stirling Highway, Crawley, WA, Australia 6009.
Plos One
|July 22, 2016
Summary
This study validated Kinect sensor software for recognizing fundamental movement skills (FMS) during active video games. The Kinect system accurately counted jumps and sidesteps, matching human rater performance but with greater efficiency.
Area of Science:
- Movement science
- Human-computer interaction
- Pediatric kinesiology
Background:
- Active video games increase energy expenditure but lack detailed movement analysis.
- Fundamental Movement Skills (FMS) are crucial for child development.
- Classifying FMS during gameplay is essential for understanding active gaming's impact.
Purpose of the Study:
- To validate motion capture software using Kinect sensor technology for recognizing FMS.
- To compare the accuracy of Kinect-based movement recognition against human assessors.
- To assess the reliability of automated FMS classification in active video game play.
Main Methods:
- Utilized Kinect sensor motion capture technology to record movements.
- Compared Kinect Action Recognition Tool (KART) data with two human assessors' ratings.
- Analyzed jump and sidestep movements during five minutes of active video game play in 43 children.
- Assessed inter-rater reliability between human assessors and agreement between human assessors and KART.
Main Results:
- High inter-rater reliability was observed for both jump (r = 0.94) and sidestep (r = 0.87) movements.
- Excellent reliability (r = 0.84) was found between human raters and KART for jumps.
- Moderate reliability (r = 0.6983) was found for sidesteps between human raters and KART.
- KART demonstrated comparable accuracy to human raters but significantly reduced analysis time.
Conclusions:
- The Kinect sensor and KART software can reliably quantify jumps and sidesteps during active video game play.
- The system offers a time-efficient alternative to manual human assessment for FMS analysis.
- This technology holds potential for objective assessment of physical activity and skill development in active gaming contexts.
