复杂场景中多细节层次模型生成新算法

A New Algorithm of Creating Levels of Detail Model in Complicated Scene

图形渲染速度目前已成为大规模虚拟漫游场景的主要瓶颈,而降低场景复杂度是提高图形渲染速度的有效方法,目前比较常用的是细节层次(Levels of Detail,简称LOD)模型。在详细分析传统LOD模型算法的基础上,提出了模型差值信息渐增的改进思想,以避免系统的重复建模并减少了数据冗余。算法验证结果表明,改进的LOD算法可以较好的降低场景复杂度,进而提高图形渲染速度。

Nowadays, the speed of graphics rendering has become the major bottleneck to large-scale virtual roaming scene, and reducing scene complexity is an effective method to improve the rendering speed of the scenes. At present, Levels of Detail is commonly used. The improved method which is gradual growth of different model values has been put forward in this thesis, and it is based on detailed analysis the traditional LOD algorithm, in order to avoid repeating models of system and reduce data redundancy. It has been showed that the improved LOD algorithm can reduce scene complexity and improve the graphics rendering speed based on the verification results of algorithm.