Training method and device of lane line detection model, equipment and storage medium

The invention discloses a training method and device of a lane line detection model, equipment and a storage medium, and belongs to the technical field of computers. The method comprises the following steps: acquiring a point cloud aerial view of a specified point cloud map, and a direction label and a direction field value label corresponding to the point cloud aerial view; inputting the point cloud aerial view into a feature extraction network to obtain image features corresponding to the point cloud aerial view; inputting the image features and the direction labels into a lane line direction identification network to obtain a horizontal lane line point cloud and a vertical lane line point cloud; inputting the image features and the direction field value labels into a lane line direction field detection network to obtain a horizontal direction field value and a vertical direction field value of the lane line point cloud; and based on the image features, the direction labels, the direction field value labels