Ooint cloud key point extraction method based on deep learning

The invention belongs to the field of artificial intelligence and the field of automatic driving, and particularly provides a point cloud key point extraction method based on deep learning. Firstly, a large number of invalid point sets are filtered by using a ground segmentation technology in a data preprocessing stage, and the key point extraction effect and efficiency are improved; Secondly, deep learning technology is used to the point cloud key point extraction problem, GPU parallel computing is used to greatly increase the computing speed, a key point extraction model including a point cloud division module and a key point extraction network is created. The farthest point sampling node is used in the point cloud division module, the expansion theory and node judgment improve the point cloud division speed, and the attention mechanism is used in the key point extraction network to improve the key point extraction performance. In conclusion, the calculation time is not greatly increased along with extracti