嵌套二叉树实现大规模地形实时绘制

实时调入视点可见数据,快速动态地建立连续细节层次(LOD)模型是提高大规模地形绘制速度的关键技术。在数据分块组织的基础上构建了两层嵌套二叉树。通过索引数据二叉树和可见性判断,实现可见数据的实时调入;对可见数据块,通过LOD建模三角形二叉树和设计的基于包围球的节点误差评价函数,实现连续LOD模型的快速动态构建,提高大规模地形的绘制速度。实验表明文中的方法可以取得较高的帧速率和较好的绘制效果,实现了大规模地形的实时绘制。

Large Scale Terrain Real-time Rendering Using Nested Bintree

The real-time loading of terrain data in view field and building continuous Level of Details (LOD) are the key technologies for improving the rendering speed of large terrain. By organizing data block partition, two layer nested bintree were built. By traversing index bintree and visibility estimation, the real-time loading of terrain data was implemented; for the data block in view field, by traversing LOD triangle bintree and designing a node error metric based on ball, continuous LOD construction was implemented quickly and dynamically, the rendering speed of large scale terrain was improved. As a result, the algorithm can achieve a good visual effect and a high FPS, and implement large scale terrain real-time rendering.