大型虚拟场景的分割与浏览优化

浏览VRML生成的大型虚拟场景时,通常绘制的开销都很大,由此制约了场景显示的实时性,也就是难以做到顺畅浏览.实际应用中,用户不得不在模型的精确度、绘制的真实感和显示的实时性三方面做出折衷的选择.以浙江师范大学虚拟校园为例,重点探讨在尽可能保证绘制的真实感的前提下,如何对大型虚拟场景用空间格网实现分割,以及采用VRML的传感器、Inline节点与编程相结合的场景调度,以解决浏览的优化问题.

Division and Browse Optimization of Large-scale Virtual Scenes

To browse large-scale scenes from VRML usually costs a lot on render,it results in certain limitations on render timely,that means favoring browse is not easy to reach.Users have to take an eclectic option among model's precision,render reality and render timely.With the example of virtual campus scene of Zhejiang Normal University,to keep sense of drauoing as real ass possible,this paper mainly discusses how to divide large-scale virtual scenes with space grid,how to introduce scenes formulation in combination of programs and sensor,Inline Node of VRML in order to achieve optimization on browse scenes.