Elderly Exercise Application Based on Real-Time Motion Tracking Using AI
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
This study aimed to develop a customized exercise application for the elderly to enhance and maintain their health. The application uses smartphone-based real-time motion tracking technology and AI learning based on videos and images to implement an exercise motion model capable of recognizing and judging step-by-step movements. The elderly-specific exercise application should support real-time feedback and be designed to allow progressive exercises that suit the elderly's physical capabilities and range of motion. For this purpose, the Goal Post Squeeze exercise motion was selected and designed as a series of unit movements. A motion recognition model was developed based on the MoveNet pose estimation technique. The operational experiment results of the implemented exercise motion model confirmed that it recognizes and judges step-by-step data, distinguishes between correct and incorrect motions, assesses horizontal alignment, and provides real-time feedback to the user based on this information.
Abstract
This study aimed to develop a customized exercise application for the elderly to enhance and maintain their health. The application uses smartphone-based real-time motion tracking technology and AI learning based on videos and images to implement an exercise motion model capable of recognizing and judging step-by-step movements. The elderly-specific exercise application should support real-time feedback and be designed to allow progressive exercises that suit the elderly's physical capabilities and range of motion. For this purpose, the Goal Post Squeeze exercise motion was selected and designed as a series of unit movements. A motion recognition model was developed based on the MoveNet pose estimation technique. The operational experiment results of the implemented exercise motion model confirmed that it recognizes and judges step-by-step data, distinguishes between correct and incorrect motions, assesses horizontal alignment, and provides real-time feedback to the user based on this information.
참고문헌 (13)
- Switonski, Adam, "Dynamic time warping in classification and selection of motion capture data", Multidimensional Systems and Signal Processing, 30, pp.1437~1468 (2019)
- Mian, Omar S., "Gait in SWEDDs patients: comparison with Parkinson's disease patients and healthy controls", Movement Disorders, 26(7), pp.1266~1273 (2011)
- Plotnik, Meir, "A new measure for quantifying the bilateral coordination of human gait: effects of aging and Parkinson’s disease", Experimental brain research, 181, pp.561~570 (2007)
- Osgouei, Reza Haghighi, "Rehabilitation exergames: Use of motion sensing and machine learning to quantify exercise performance in healthy volunteers", JMIR Rehabilitation and Assistive Technologies, 7(2), pp.e17289~ (2020)
- Yu, Jimin, "Dynamic gesture recognition based on 2D convolutional neural network and feature fusion", Scientific Reports, 12(1), pp.4345~ (2022)
- Lei, Qing, "Learning effective skeletal representations on RGB video for fine-grained human action quality assessment", Electronics, 9(4), pp.568~ (2020)
- Kendall, Alex, "Posenet: A convolutional network for real-time 6-dof camera relocalization", Proceedings of the IEEE international conference on computer vision, pp.2938~2946 (2015)
- U.S. Department of Health and Human Services, "Physical activity guidelines for Americans 2nd edition", ~ (2018)
- S..G. Chung, "Baeknyeon movement", pp.64~97 (2019), Artisanseoul
- National Health Insurance Service, "National Health Insurance Statistical Yearbook" (2022)
- "MoveNet"[URL]
- "MoveNet"[URL]