基于视频的人工智能分析系统在引体向上运动教学及考核中的初步应用

目的 从运动医学的角度,探讨基于视频的人工智能分析系统在引体向上动作教育及考核中的应用.方法 本研究通过采集了51例引体向上的动作视频样本,同时邀请空军军医大学体育教研室专家对这些样本进行了人工计数与标准化评估,人工计数和评估将作为真值来评估本研究提出的引体向上智能考核与评估系统的性能.使用人工智能算法,智能分解、分析...

Preliminary application of video-based artificial intelligence analysis system in pull-up exercise teaching and assessment

Objective To investigate the application of video-based artificial intelligence analysis system in the education and assessment of pull-up exercise from the perspective of sports medicine.Methods Pull-up action video samples of 50 participants were collected. At the same time, experts from the Physical Education were invited to manually count and standardize these samples. The manual evaluation was used as the ground truth to evaluate the performance of the proposed intelligent evaluation system. Artificial intelligence algorithms were used to decompose and analysis the video samples, thus the counting requirements were realized. At the same time, each pull-up movement was analyzed to help students find the key points of non-standard movements.Results The evaluation count error of the proposed intelligent evaluation system was 0.24, and the average error of standardized evaluation was 1.27.Conclusion The proposed artificial intelligence analysis system has preliminarily realized the intelligent counting and standardized assessment of the pull-up exercise.