Map data labeling method, device, equipment and storage medium

Abstract

Embodiments of the present application provide a map data labeling method, device, device, and storage medium, by acquiring point cloud data of a target scene; based on the point cloud data, fitting the ground, and generating a two-dimensional map of the target scene; Based on the first position of the ground surface element in the two-dimensional map, determine the second position of the ground surface element on the ground, wherein the ground surface element refers to a road sign drawn on the road surface; acquire based on the point cloud data the three-dimensional data of the surface element at the second position, and mark the three-dimensional data of the surface element on a three-dimensional map. The technical solutions provided by the embodiments of the present application can improve the labeling speed of map data, especially three-dimensional map data.

Description

technical field [0001] The embodiments of the present application relate to the field of computer technology, and in particular, to a method, device, device, and storage medium for labeling map data. Background technique [0002] Autonomous driving technology needs to rely on high-precision maps. Vehicles can drive based on road information in high-precision maps. Therefore, the road information in high-precision maps needs to be very comprehensive and accurate. This leads to a very high complexity of high-precision maps, including a lot of data on various road elements (such as traffic lights, road lines, etc.), making the production of high-precision maps very difficult and very low in production efficiency. [0003] At present, in the process of generating high-precision maps, road elements are usually directly identified and extracted from the point cloud, and then the three-dimensional data of the road elements in the point cloud are marked on the high-precision map. In...

AI Technical Summary

Problems solved by technology

This leads to a very high complexity of high-precision maps, including a lot of data on various road elements (such as traffic lights, road lines, etc.), making the production of high-precision maps very difficult and very low in production efficiency

[0003] At present, in the process of generating high-precision maps, road elements are usually directly identified and extracted from the point cloud, and then the three-dimensional data of the road elements in the point cloud are marked on the high-precision map. In this way, based on point Cloud identification processing is more complex, time-consuming and less efficient

Images