Application and Research of Basketball Footwork Supported by Intelligent Edge Cloud Computing and Deep Learning Unsupervised Transfer Method

Basketball is a comprehensive physical sport in which footwork migration is the key point. To explore the current teaching status of basketball footwork mobile, Deep Learning (DL) and unsupervised transfer methods are combined to extract the footwork movement characteristics of basketball players for data analysis and research. At the same time, to effectively analyze the feature data of the collected player footwork, intelligent edge cloud computing is used to carry out advanced processing on the extracted data features, and Convolutional Neural Networks (CNNs) are employed to perform transfer learning on the extracted target domain samples. Feature extraction is performed on the data, and the impending action of a basketball player is predicted. Meanwhile, the unsupervised human action transfer method is studied to provide new ideas for basketball footwork action series data modeling. Finally, the theoretical framework based on intelligent edge cloud computing and DL unsupervised transfer method is summarized. Its principle is explored and applied in the teaching of basketball footwork. The results show that: (1) The converged convolutional network and classification network parameters can predict players’ movement trajectories. (2) Compared with the existing supervised learning methods on synthetic datasets, unsupervised training using network data dramatically increases the variety of actions during training. (3) The classification accuracy of the transfer learning method is high, and it can be used for the different basketball footwork in the corresponding stage of the court.