点云驱动的计算机图形学综述  

Survey on Points-Driven Computer Graphics

作　　者：伍龙华[1] 黄惠[1] 

机构地区：[1]中国科学院深圳先进技术研究院,深圳518055

出　　处：《计算机辅助设计与图形学学报》2015年第8期1341-1353,共13页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61202224;61331018);国家"八六三"高技术研究发展计划(2013AA01A604);深圳市可视计算与可视分析重点实验室(CXB201104220029A);深圳市基础研究项目(JCYJ20140901003938994)

摘　　要：点云是分布在三维空间中的离散点集,也是对物体表面信息的离散采样.三维扫描技术的迅速发展使得点云数据的获取更加简单方便,而点云驱动的计算机图形学在逆向工程、数字城市、文物保护、智能机器人等领域日益彰显其广阔的应用前景.文中简要介绍了近年来计算机图形学围绕点云在数据获取、采样处理、渲染表示和几何建模4个方面的主要工作,展示了这一新兴领域的研究现状和发展趋势,并探讨了未来有潜力的研究方向.A point cloud is a set of discrete points distributed in 3D space. The development of 3D scanning technology makes it possible to rapidly capture and sample the shapes of objects as point clouds. Points-driven computer graphics is hence now widely applied in many fields, such as reverse engineering, digital city, cultural relics preservation and robotics. This survey briefly introduces the recent computer graphics research based on point clouds in four aspects： capturing, resampling, representation and reconstruction. It provides the readers a big picture on this new rising field, showing state-of-the-arts, study trends and potential research directions in the future.

关 键 词：点云 三维扫描 配准 骨架提取 渲染 曲面重建 动态建模 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]