Multi-view point splatting
Multi-view point splatting
复制标题
多视点泼溅
DOI：
--
复制
发表时间：
2006
期刊：
影响因子：
--
通讯作者：
R. Pajarola
中科院分区：
文献类型：
--
作者：
T. Hübner;Yanci Zhang;R. Pajarola
当前立体和多视图可视化的根本缺点是必须执行多遍渲染(每个视图一遍)和随后的图像合成+遮罩以生成多个立体视图。因此，渲染时间一般随着视点数量的增加而线性增加。本文提出了一种基于延迟混合的多视点拼接方法。我们的方法利用现代图形处理单元(GPU)的可编程性，在一次渲染过程中渲染多个立体视图。这些视图直接在GPU上计算，包括亚像素波长选择视图。我们精确地描述了我们的算法，并提供了它的实现细节。实验结果表明，与标准的多遍算法相比，本文的多视点拼接算法具有更好的性能优势。
The fundamental drawback of current stereo and multi-view visualization is the necessity to perform multi pass rendering (one pass for each view) and subsequent image composition + masking for generating multiple stereo views. Thus the rendering time increases in general linearly with the number of views.In this paper we introduce a new method for multi-view splatting based on deferred blending. Our method exploits the programma-bility of modern graphic processing units (GPUs) for rendering multiple stereo views in a single rendering pass. The views are calculated directly on the GPU including sub-pixel wavelength selective views. We describe our algorithm precisely and provide details about its implementation. Experimental results demonstrate the performance advantage of our multi-view point splatting algorithm compared to the standard multi-pass approach.