Point cloud video up-sampling method based on feature fine tuning

The invention discloses a feature fine tuning-based point cloud video up-sampling method, which comprises a model training process and a model reasoning process, and is characterized in that in the model training process, a sparse point cloud video subjected to sparse sampling in a training data set is used as the input of a model, and an output point cloud video is obtained after up-sampling; the model reasoning process comprises the steps of inputting any sparse point cloud video into a trained model, and obtaining a dense point cloud video corresponding to the input sparse point cloud video after up-sampling; the up-sampling process comprises a feature extraction stage, a feature fine adjustment stage and a point cloud reconstruction stage. According to the method, the characteristic of high similarity of adjacent frames of the point cloud video is utilized, a feature fine adjustment method is used, the calculation amount of a feature extraction part with the maximum calculation amount is reduced, the mode