Yumei Shen
Research Subjects
Yumei Shen has a total of 3 co-authors. Here are the top co-authors by number of shared publications:
- Shun Nian Luo
School of Information Science and Technology, Sanda University, Pudong District, Shanghai 201209, P. R. China
1 shared publication - Wu-Shiun Tao
Department of Ball Sport, National Taiwan University of Sport, No. 16, Sec. 1, Shuangshi Rd., North Dist., Taichung City 404401, Taiwan, R.O.C.
1 shared publication - Zhi-Shang Zhang
Department of Physical Education, Guangdong Medical University, Songshan Lake Science and Technology Industrial Park, Dongguan 523808, Guangdong Province, P. R. China
1 shared publication
Latest Publications
- 31 March 2026
- Published by World Scientific Pub Co Pte Ltd in Journal of Mechanics in Medicine and Biology
- Vol. 26 (04)
THE DETECTION AND ANALYSIS OF THE MOTION TRAJECTORY AND SAFE PUNCHING OF BOXING BASED ON CNN
This study employs MediaPipe posture detection technology for motion capture, focusing on observing the movement of several key points of the right arm. By comparing the data model of the standard action, it can determine whether the current action is within the safe range and give judgment tips. The test includes OpenCV image processing and video analysis, Matplotlib data visualization combined with NumPy numerical calculation to realize action evaluation, providing feedback according to data comparison, convolutional neural network model training, and data management core function modules. Results indicate that the system performs well in judging action normativity, with errors within the normal range; error is basically within the normal range. This could help prevent sports injuries. This portable smart monitoring system uses a regular camera to record movements and then uploads them to a computer for analysis. The advantage is that it is convenient and practical.
- 31 March 2026
- Published by World Scientific Pub Co Pte Ltd in Journal of Mechanics in Medicine and Biology
- Vol. 26 (04)
THE DETECTION AND ANALYSIS OF THE MOTION TRAJECTORY AND SAFE PUNCHING OF BOXING BASED ON CNN
This study employs MediaPipe posture detection technology for motion capture, focusing on observing the movement of several key points of the right arm. By comparing the data model of the standard action, it can determine whether the current action is within the safe range and give judgment tips. The test includes OpenCV image processing and video analysis, Matplotlib data visualization combined with NumPy numerical calculation to realize action evaluation, providing feedback according to data comparison, convolutional neural network model training, and data management core function modules. Results indicate that the system performs well in judging action normativity, with errors within the normal range; error is basically within the normal range. This could help prevent sports injuries. This portable smart monitoring system uses a regular camera to record movements and then uploads them to a computer for analysis. The advantage is that it is convenient and practical.