【24h】

GPU Acceleration of Improved Particle-based Volume Rendering for Irregular-Grid Volume Data

机译:GPU加速,改进了基于不规则网格体数据的基于粒子的体绘制

获取原文

摘要

In this paper,we apply an improved particle-based volume rendering (PBVR) technique for previewing large irregular-grid volume data using current programmable GPU architecture.This technique allows opaque and emissive particles rendering of translucent volumes without visibility sorting.In our new GPU acceleration of PBVR,we provide a scalability feature by developing cell-by-cell particle generation during rendering,so that the user can alternatively switch to particle pre-generation mode for higher speed rendering.We also reduce the memory cost required for storing all sub-pixel values by proposing a pixel-superimposing technique for the large sub-pixel level.By adjusting the repeat level we achieved a very smooth level of detail (LOD) control to exchange quality with speed.As another improvement,we enhanced the image quality by introducing a post-rendering filter,which degrades the speckle artifact derived from particle sampling.Our work demonstrates a full detail rendering rate from 5 to 10 fps for irregular-grid volume data with mega-scale amounts of cell on NVIDIA GEFORCE 8800GTS.
机译:在本文中,我们应用了一种改进的基于粒子的体绘制(PBVR)技术,以使用当前的可编程GPU架构预览大型不规则网格体数据。该技术允许对半透明体积的不透明和自发光粒子进行渲染,而无需进行可见性排序。 PBVR的加速,我们通过在渲染期间开发逐个单元的粒子生成来提供可伸缩性功能,以便用户可以替代地切换到粒子预生成模式以实现更快的渲染。我们还降低了存储所有子对象所需的内存成本通过针对大的子像素级别提出像素叠加技术来提供像素值。通过调整重复级别,我们实现了非常平滑的细节级别(LOD)控制,以实现质量与速度的交换。作为另一项改进,我们提高了图像质量通过引入后期渲染滤镜,该滤镜可降低源自粒子采样的斑点斑点。我们的工作证明了完整的细节渲染速率在NVIDIA GEFORCE 8800GTS上以5至10 fps的速度处理具有格状单元的不规则网格体数据。

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号