首页 | 本学科首页   官方微博 | 高级检索  
     

基于BRDF 和GPU 并行计算的全局光照实时渲染
引用本文:王 芳,秦磊华. 基于BRDF 和GPU 并行计算的全局光照实时渲染[J]. 图学学报, 2016, 37(5): 583. DOI: 10.11996/JG.j.2095-302X.2016050583
作者姓名:王 芳  秦磊华
作者单位:1. 郑州升达经贸管理学院信息工程系,河南 郑州 451191;2. 华中科技大学计算机科学与技术学院,湖北 武汉 430074
摘    要:基于光线追踪,将屏幕图像像素分解为投射光线与场景对象交点面片辐射亮度和纹理贴图的合成,每个面片的辐射亮度计算基于双向反射分布函数(BRDF)基的线性组合,并通过图形处理器(GPU)处理核心并行绘制进行加速,最后与并行计算的纹理映射结果进行合成。提出了一种基于BRDF 和GPU 并行计算的全局光照实时渲染算法,利用GPU 并行加速,在提高绘制效率的前提下,实现动态交互材质的全局光照实时渲染。重点研究:对象表面对光线的多次反射用BRDF 基的线性组合来表示,将非线性问题转换为线性问题,从而提高绘制效率;利用GPU 并行加速,分别计算对象表面光辐射能量和纹理映射及其线性组合,进一步提高计算效率满足实时绘制需求。

关 键 词:全局光照  图形处理单元  双向反射分布函数  渲染方程  并行计算  

Global Illumination Real-Time Rendering Based on BRDF and#br#GPU Parallel Computing
Wang Fang,Qin Leihua. Global Illumination Real-Time Rendering Based on BRDF and#br#GPU Parallel Computing[J]. Journal of Graphics, 2016, 37(5): 583. DOI: 10.11996/JG.j.2095-302X.2016050583
Authors:Wang Fang  Qin Leihua
Affiliation:1. Department of Information Engineering, Zhengzhou Shengda College of Economics & Trade Management, Zhengzhou Henan 451191, China;2. School of Computer Science & Technology, Huazhong University of Science & Technology, Wuhan Hubei 430074, China
Abstract:While raytracing, the screen image pixel is decomposed into the combination of radianceand texture of the patches, created as scene objects intersect with the casting ray. The radiance of eachpatch is calculated at the linear combination of the bases of bi-directional reflectance distributionfunction (BRDF), and able to be accelerated by graphics processing unit (GPU) parallel rendering.This paper presents a global illumination rendering algorithm based on BRDF and GPU parallelcomputation. With GPU parallel acceleration, through improving the efficiency of rendering, thealgorithm achieves global illumination real-time rendering of the scene including dynamic interactivematerial. The key research: object surface’s multiple reflection characteristic is represented by linearcombination of the basis of BRDF, so transforming the nonlinear problem to a linear one, thusimprove the rendering efficiency. With GPU parallel acceleration, the algorithm calculates the objectsurface’s radiation energy and texture mapping and their linear combination, further improving theefficiency of rendering to meet the requirement of real-time.
Keywords:global illumination,graphics processing unit,bidirectional reflectance distribution function  rendering equation,parallel computing,
本文献已被 CNKI 等数据库收录!
点击此处可从《图学学报》浏览原始摘要信息
点击此处可从《图学学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号