基于连续HLOD的大规模场景快速绘制算法

大规模场景的快速绘制是虚拟现实技术重要的研究课题之一.为了加速场景的绘制,一般采用层次细节模型和可见性裁剪方法,但是现有算法在处理大规模场景时存在着局限性.本文提出了一种新的大规模场景快速绘制算法,该算法在场景层次划分的基础上,利用拓扑结构可变的网格简化方法为场景层次计算连续的分层层次细节模型（HLOD）;然后在实时绘制阶段,对场景分层层次细节模型进行视点相关的全局和局部细化,并结合快速有效的视域裁剪,从而大大加速了场景绘制速度.实验结果表明该算法是简单有效的,并且算法还可以进一步扩展到外存方式.

Fast Rendering Algorithm for Large-scale and Complex Scene Based on Continuous HLOD

Real-time rendering of large-scale and complex scenes is one of the important subjects in virtual reality technology.This paper presents a new fast rendering algorithm for large-scale and complex scenes.In the preprocessing phase,the algorithm partitions the scene in hierarchy and compute continuous hierarchical levels of detail(HLOD) for each hierarchical node using a mesh simplification method which supports topological structure modification.Then at run-time,the algorithm uses the hierarchy for coarse global refinement,and uses the continuous HLOD for fine local refinement.The view frustum culling can also be efficiently implemented.As the result,the method can greatly accelerate the rendering speed.Experiment results show that it is very efficient,practical and easy to implement.Moreover,the method can also be implemented in out-of-core mode.