摘要
近年来,随着3D扫描设备的升级换代,点云数据的获取愈发便捷。相较于图像,点云数据包含更为丰富的空间信息和几何信息,但由于点云数据具有无序性和非结构化特征,导致点云语义分割方法发展受限。针对以上问题,分别从传统方法和基于深度学习的方法对点云语义分割算法进行总结与介绍。
In recent years,with the upgrading of 3D scanning equipment,the acquisition of point cloud data is becoming more convenient.Compared with images,point cloud data contains more abundant spatial and geometric information.However,due to the disordered and unstructured characteristics of point cloud data,the development of point cloud semantic segmentation methods is limited.Aiming at the above problems,the paper summarizes and introduces the point cloud semantic segmentation algorithm from the traditional method and the method based on deep learning.
作者
冯站银
FENG Zhanyin(School of Advanced Manufacturing,Fuzhou University,Quanzhou 362200,China;Chinese Academy of Sciences,Quanzhou Institute of Equipment Manufacturing,Quanzhou 362216,China)
出处
《电视技术》
2023年第3期140-143,148,共5页
Video Engineering
关键词
点云
语义分割
深度学习
point cloud
semantic segmentation
traditional method
deep learning
