Method and device for processing three-dimensional point cloud and storage medium

Abstract

The invention discloses a three-dimensional point cloud processing method and device and a storage medium. The method comprises the steps that any one of at least two processing modules receives a three-dimensional point data set, the three-dimensional point data set comes from all levels of processing modules before the processing module, and area feature data of at least two space areas in the three-dimensional point data set are obtained respectively, the regional feature data of the at least two spatial regions are gathered to obtain the feature data of the three-dimensional point data set, so that the complexity of the calculation process of the three-dimensional point cloud is reduced, the calculation speed is improved, and if the processing module is the processing module of the middle layer in the processing model, the processing efficiency is improved. The feature data of the three-dimensional point data set are respectively output to each layer of processing module after the processing module, and if the processing module is the last layer of processing module in the processing model, the feature data of the three-dimensional point data set is used as geometric representation corresponding to the three-dimensional point cloud, so that geometric information missing does not occur, and the accuracy of the geometric representation is improved. And the accuracy of the data obtained by the three-dimensional point cloud is improved.

Description

technical field [0001] The present application relates to the field of data processing, in particular to a method, device and storage medium for processing three-dimensional point clouds. Background technique [0002] A 3D point cloud refers to a collection of multiple points in a 3D space, and each point in the 3D point cloud can be regarded as a 3D vector. The 3D point cloud can represent the semantic features of the object such as the shape and color of the object in the 3D space through the collection of points. Based on this, processing the 3D point cloud to obtain data such as the shape of the object provides a parameter basis for classifying and identifying objects in the 3D space. [0003] Existing methods for processing 3D point clouds include methods based on multi-view mapping, methods based on voxel conversion, and direct modeling methods based on deep learning. In the process of processing the 3D point cloud, the method will cause different degrees of geometri...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the above-mentioned technical problems in the prior art, the embodiment of the present application proposes a method, device and storage medium for processing 3D point clouds, so as to solve the complicated calculation process of these methods for processing 3D point clouds, and from these methods respectively The geometric information extracted from the 3D point cloud has the problem of missing information in different dimensions

Images