大规模复杂场景的可见性问题研究  

Research on Visibility for Large-Scale and Complex Scenes

作　　者：普建涛[1] 查红彬[1] 

机构地区：[1]北京大学计算机科学系视觉与听觉信息处理国家重点实验室,北京100871

出　　处：《计算机研究与发展》2005年第2期236-246,共11页Journal of Computer Research and Development

基　　金：国家自然科学基金重点项目 (60 3 3 3 0 10 );中国博士后科学基金项目 (2 0 0 3 0 3 3 0 78)

摘　　要：大规模复杂场景的快速绘制技术是虚拟现实、实时仿真以及三维交互设计等许多重要应用的底层支撑技术 ,也是诸多研究领域面临的一个基本问题 作为其中一个技术解决手段 ,可见性问题在近年来得到了高度重视并取得了一系列研究成果 通过对相关可见性算法进行分析与总结 ,阐述了可见性问题的研究内容 ,提出了方法的优劣判别标准 ,给出了可见性判断算法所应包含的基本组成部分和步骤 。The real-time rendering technique of large-scale and complex scenes is the basis of many important applications such as virtual reality, real-time simulation, and 3D interactive design. As one way to accelerate the rendering speed, the research on visibility has gained great attention in recent years. Its aim is to discard invisible geometric objects as many as possible based on given viewing points so that complex scenes or objects can be rendered or transferred through Internet at real time. In this paper, a survey related to this topic is presented. First, some typical visibility algorithms are analyzed in detail. Second, some topics related to visibility research are introduced and the necessary components and steps that a visibility algorithm needs are described. Third, some criteria to evaluate the visibility methods are proposed. Finally, some key techniques that need further research efforts are pointed out.

关 键 词：可见性 剪裁 大规模复杂场景 实时漫游 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]