Instant Level-of-Detail
Instant Level-of-Detail
复制标题
即时细节级别
DOI：
--
复制
发表时间：
2011
期刊：
影响因子：
--
通讯作者：
M. Guthe
中科院分区：
文献类型：
--
作者：
Nico Grund;Evgenij Derzapf;M. Guthe
高度详细的模型通常用于计算机游戏和其他交互式渲染应用程序中。在这种情况下，静态细节级别经常用于实现实时帧速率。虽然这是提高渲染性能的简单解决方案，但需要将额外的几何体存储并加载到图形内存中。这在在线应用程序中尤其成问题，因为数据需要通过可能很慢的连接传输。另一方面，消费级计算机通常配备有可用于通用并行计算的图形卡。基于这一观察，我们提出了一个高质量的并行网格简化算法的基础上的二次误差度量。简化性能可以与从本地硬盘加载额外网格所需的时间竞争。
Highly detailed models are commonly used in computer games and other interactive rendering applications. In this context, static levels-of-detail are frequently used to achieve real-time frame rates. While this is a simple solution to improve the rendering performance, the additional geometry needs to be stored and loaded into graphics memory. This is especially problematic in online applications, where the data needs to be transmitted over a possibly slow connection. On the other hand, consumer level computers are usually equipped with a graphics card that can be used for general purpose parallel computing. Based on this observation, we propose a high-quality parallel mesh simplification algorithm based on the quadric error metric. The simplification performance can compete with the time required to load additional meshes from a local hard disk.