针对传统阴影绘制算法在构建阴影图时计算量大的问题,提出一种基于BVH的自适应阴影图算法。该算法能够对三维场景中的几何体自下而上地快速构建基于AABB包围盒的BVH空间索引结构,并对SAH代价函数加以改进,构建包含“几何体组-面掩码”对的列表,可以自适应地剔除与对应光源无关的几何体,减少阴影图构建时的计算量。实验结果表明,与传统算法相比,本文方法能够有效加快阴影图的构建速度,提高三维场景的阴影绘制效率。

In order to overcome the problem of large amount of computation in the shadow map construction by traditional algorithm, an adaptive shadow map algorithm based on BVH was proposed. In this algorithm, the geometry in 3D scene could be organized by BVH spatial index structure based on the AABB bounding box rapidly, and the SAH cost function was optimized, then a list containing “geometry group-face mask” pairs was constructed, which could adaptively cull the geometry that was independent of the corresponding light source, and the amount of computation in the construction of shadow map could be reduced. The experiment results show that, compared with the traditional algorithm, the construction time of shadow map was decreased, and the efficiency of shadow rendering was improved.