点云编码综述  

A Review of Point Cloud Compression

作　　者：李厚强[1] 李礼[1] 李竹 LI Houqiang;LI Li;LI Zhu(University of Science and Technology of China,Hefei 230027,China;University of Missouri-Kansas City,Kansas City 64110,USA)

机构地区：[1]中国科学技术大学,中国合肥230027 [2]美国密苏里大学堪萨斯分校,美国堪萨斯城64110

出　　处：《中兴通讯技术》2021年第1期5-9,共5页ZTE Technology Journal

基　　金：国家自然科学基金(62021001);中国科学技术大学引进人才科研启动经费(KY21000000108)。

摘　　要：点云编码是支撑点云广泛应用的关键技术之一,是近期技术研究和标准化领域的热点。对点云几何信息和属性信息编码技术演进进行了回顾,并针对稠密点云和稀疏点云的几种典型编码方法的编码效率进行了比较。未来点云编码研究将集中于利用帧间预测去除动态点云的不同帧之间的相关性,以及端到端点云编码、任务驱动的点云编码等方面。3D point cloud compression is one of the key technologies supporting the widespread use of point clouds.Recently,it is one of the focuses for both research and standardization groups.The latest advance of the compression technologies for both the 3D point cloud geometry and attribute information is reviewed.Compression efficiencies of several typical compression technologies for both the 3D dense and sparse point clouds are compared.In the future,more studies will focus on inter-frame prediction to exploit the correlations between different frames in 3D dynamic point clouds,end-to-end point cloud compression,and task-driven point cloud compression.

关 键 词：3D点云编码 几何信息编码 属性信息编码 

分 类 号：TN919.81[电子电信—通信与信息系统]