摘要
针对已有的软阴影映射技术允许在使用一张阴影贴图的条件下对复杂的动态场景实时渲染出比较真实的软影,而在处理大面积软影时算法的执行效率不高的问题,通过创建一种新的边界存储结构,提出了一种快速软影生成算法.该算法是一种3遍算法:第一遍从光源中心计算场景的深度图;第二遍采用边缘提取算子对深度图进行滤波,找到所有可能的边界点;第三遍查询可能的边界点,计算其对应的轮廓边,再反投影轮廓边到光源区域计算轮廓边的遮挡贡献,最后累积所有边界点的遮挡贡献得到着色点的可见性,生成真实的软影效果.实验结果表明,与现有算法相比,文中算法在现有图形硬件中执行效率更高.
The recent soft shadow mapping technique allows the rendering in real-time of convincing soft shadows on complex and dynamic scenes using a single shadow map. While attractive, this method becomes extremely expensive when dealing with large penumbrae. This paper proposes a new edge data structure to remove these limitations and hence providing an efficient and practical technique for soft shadow generation, which is an algorithm with three passes. First, we generate the depth map from the light view point. It is then followed by the second pass to extract the silhouette edge using our new area light edge operator. In the rendering pass, for each fragment we query all edge points in kernel area, find its corresponding silhouette edge, and back-project them onto the 2D light source domain, in which the occluded area is radially integrated. Finally, we use the visibility of the fragment to generate realistic soft shadow. Experimental results show that the algorithm can be used to achieve good results with the current graphics hardware.
出处
《计算机辅助设计与图形学学报》
EI
CSCD
北大核心
2011年第8期1285-1293,共9页
Journal of Computer-Aided Design & Computer Graphics
基金
国家"八六三"高技术研究发展计划(2007AA01Z317)
关键词
软影
面光源
轮廓边提取
像素着色器
soft shadow
area light
silhouette extraction
pixel shader