Method and apparatus for marking three-dimensional point cloud based on fusion voxel

Abstract

Embodiments of the present invention provide a three-dimensional point cloud marking method and apparatus based on a fusion voxel. The method comprises the following steps: the data set of the three-dimensional point cloud is voxelized and voxel features in the voxels are extracted based on the processing results to form a first voxel feature matrix; the first voxel feature matrix is used as the input of the three-dimensional convolution neural network to calculate the multi-scale feature of the voxel, and the multi-scale feature is fused in series to obtain the second voxel feature matrix. The first voxel feature matrix is used as the input of the three-dimensional convolution neural network to calculate the multi-scale feature of the voxel. Based on the feature interpolation algorithm, the voxel features in the second voxel feature matrix are extended to the points in the three-dimensional point cloud data set to obtain the point cloud feature matrix. The feature matrix of point cloud is inputted into the multilayer perceptron to mark the attributes of three-dimensional point cloud. The invention can realize fine classification and recognition point by point, so as to further improve the performance of point cloud marking.

Description

technical field [0001] The present invention relates to the technical field of data labeling, in particular to a fused voxel-based three-dimensional point cloud labeling method and device. Background technique [0002] With the widespread application of 3D sensors (such as LiDAR, Microsoft Kinect, ASUS Xtion, etc.) in various fields (such as mobile robots, autonomous driving, remote sensing interpretation, virtual reality, augmented reality, and battlefield situational It is also growing significantly, and 3D point cloud marking, as an important means of processing 3D data, refers to identifying the category attributes of each point in the observed scene point cloud data, and assigning a unique category to each point Labels, such as buildings, roads, or cars. However, the traditional semantic labeling based on voxel convolutional neural network requires that all points in the voxel share the same semantic label, and requires regular data input, and the labeling results obta...

AI Technical Summary

Problems solved by technology

However, the traditional semantic labeling based on voxel convolutional neural network requires that all points in the voxel share the same semantic label, and requires regular data input. At the same time, the labeling result obtained is also a rough labeling result at the voxel level.

Images