Dynamic Tracking Optimization of Basketball Trajectory Based on Graph Neural Network

The design of basketball trajectory dynamic tracking can improve the basketball hit rate. An optimization method of basketball trajectory dynamic tracking based on graph neural network is proposed in this paper. A sports vision acquisition model is established by triangulation and image recognition, which records the player's trajectory information at 25 frames per second, and chooses basket attack or middle-distance shooting according to personal ability and field situation to complete the attack. According to the observation of the occurrence process of events, effective frames are selected from each event for visualization, and the position of the end point is added to the trajectory image to strengthen the marking and implementation tactics. The graph neural network model is used to realize the dynamic prediction of basketball trajectory, and all dimensional features of a single sample are normalized. And the dynamic tracking of basketball trajectory is realized according to the prediction results of the graph neural network model. The simulation results show that this method can effectively transform the basketball track sequence data into a track map, strengthen the interactive ability of feature information between the court line and the player's track, and realize the dynamic tracking and recognition of the track.