利用可编程GPU实现大规模地形场景的高性能漫游*

对已有算法进行了综述，并针对数据动态调度、自适应网格模型的生成以及数据的组织与数据裁剪等方面进行了研究并提出改进方法，设计了一种基于GPU编程实现的大规模地形场景的实时绘制与漫游算法。利用GPU端完成地形网格更新、地形块的自动选取、高度图和纹理图采样等大部分计算工作，大大减轻了CPU端的计算负载。实验表明，该算法实现简单，内存开销较少，有效提高了地形绘制的效率，适于大规模地形场景的实时高效漫游。

High performance navigation of large-scale terrain based on GPU programming

This paper studied previous algorithms, proposed several improvement methods related to data dynamic scheduling, adaptive mesh model generating, data organization and data clipping, and designed a high performance technique for real-time rendering and roaming of large-scale terrain environment based on GPU programming. GPU undertaked most of the computation such as terrain mesh updating, automatically selecting of terrain blocks, terrain blocks sampling and texture blocks sampling. Experimental results show that GPU programming is an effective way to improve the method, the algorithm is easy to implement, has a low memory expenses, and achieves high performance on real-time roaming of large-scale terrains.