Intelligent Motion Detection Software based on Efficient Spatiotemporal Graph Convolution Algorithm

Abstract

Exercise and dietary habits are important ways to maintain human health, and many people hope to ensure their own physical health through reasonable exercise and maintenance methods. Sports health analysis algorithms serve this purpose, and most intelligent motion detection software is also designed based on this algorithm. This article suggested that the performance of this algorithm can be improved through efficient spatiotemporal graph convolution, which enhanced its ability to collect data from various sources, analyze user physical conditions and exercise habits, and provide reasonable suggestions. This article also verified through comparative experiments that the motion health analysis algorithm based on efficient spatiotemporal graph convolution had stronger prediction accuracy than a single motion health analysis algorithm. The average prediction accuracy of these two algorithms was 92.93% and 84.81%, respectively. The experimental results fully demonstrated that the method proposed in this paper was very suitable for assisting in the construction of sports health analysis algorithms.