优化的GPU体绘制  

Optimizing GPU Volume Rendering

作　　者：刘云伍[1] 孙滕[1] 王刚[1] 

机构地区：[1]兰州交通大学电子与信息工程学院

出　　处：《兰州交通大学学报》2011年第1期27-32,共6页Journal of Lanzhou Jiaotong University

基　　金：甘肃省自然科学基金(08032JZA015);甘肃省科技攻关项目(0708GKCA047);兰州市科技计划项目(2010-1-6)

摘　　要：体绘制方法采用了GPU在现成的硬件上就能提供高性能的能力.本文讨论了在执行基于GPU的体绘制的时候图象硬件中出现的多种不同的瓶颈问题,提出一种新的策略,在不损害图象质量的情况下,这个策略可以平衡在特定瓶颈中的加载问题.介绍了双阶段的空间跳过,第一阶段应用于对一个半常规的网格进行分块,第二阶段使用八叉树来获得更好的间隔.同时,对数据块使用了提前光线终止的技术,并分析了两个阶段如何解决瓶颈问题和为了特定的硬件管道而如何被调节的.Volume rendering methods employ GPU capabilities offering high performance to off-the-shelf hardware.In this article,the various bottlenecks found in the graphics hardware are expounded when performing GPU-based volume rendering.A new strategy is presented to balance the load on the identified bottlenecks,without compromising the image quality.The strategy introduces a two-staged space-skipping,whereby the first stage applies bricking to a semi-regular grid,and the second stage uses octrees to reach a finer granularity.Additionally,the early ray termination is applied to the bricks.How the two stages address the individual bottlenecks,and how they can be tuned for a specific hardware pipeline are demonstrated.The described method takes into account that the rendered volume may exceed the available texture memory.

关 键 词：体可视化 直接体绘制 纹理分片 层次的绘制 GPU 

分 类 号：TP317[自动化与计算机技术—计算机软件与理论]