计算机工程与应用2025，Vol.61Issue(11)：22-30,9.DOI:10.3778/j.issn.1002-8331.2408-0033
深度学习的点云语义分割方法综述
Survey on Deep-Learning-Based Point Cloud Semantic Segmentation
摘要
Abstract
Point cloud semantic segmentation assigns semantic labels to each point in the point cloud to achieve the seg-mentation of different objects in the scene,which is the foundation for scene understanding.In recent years,with the development of deep learning technology,the combination of deep learning and point cloud semantic segmentation meth-ods has improved the processing efficiency and segmentation accuracy,demonstrating excellent performance,and has been widely used in many fields such as transportation,medicine,architectural design,virtual reality,etc.Based on the review of the development history of point cloud semantic segmentation,this paper classifies and summarizes the existing research,analyzes related datasets and evaluation metrics,and compares the performance of existing methods.Finally,the paper highlights the deficiencies of existing research and looks forward to the future development directions.关键词
深度学习/点云/语义分割/语义标签/计算机视觉Key words
deep learning/point cloud/semantic segmentation/semantic labels/computer vision分类
计算机与自动化引用本文复制引用
蔡子悦,袁振岳,庞明勇..深度学习的点云语义分割方法综述[J].计算机工程与应用,2025,61(11):22-30,9.基金项目
江苏省教育科学规划课题立项(B/2023/01/96). （B/2023/01/96）
