Precise Positioning and Calibration of Sports Teaching Movements Based on Artificial Intelligence Deep Learning Technology

Abstract

There are various complex movements in physical education and training. Therefore, it is difficult to accurately judge wrong movements by relying solely on traditional contour detection methods and cause athletes to be unable to correct movements in time. However, the application of three-dimensional modeling and detection methods can strengthen computer vision image analysis and reconstruction processing, and improve the accuracy of wrong action judgments. Based on artificial intelligence deep learning technology, this paper designs a multi-target detection algorithm, and uses a multi-target tracking algorithm fused with spatial color histograms to achieve multi-target tracking prediction. Moreover, this paper combines the monocular ranging algorithm to design a set of real-time and high-precision sports action recognition system. In addition, this paper designs experiments to verify the performance of this system. The research results show that the performance of the model constructed in this paper is good.