基于信息熵的蒙特卡罗全局光照的自适应抽样

在真实感图形生成领域里，蒙特卡罗方法是计算整体光照问题的极佳选择。但是，在用基于蒙特卡罗的全局光照算法生成的图像中，当没有足够多的采样量的时候，存在大量的噪声。自适应抽样方法是减少这种噪声的一种很好的方法。该文提出了一种新的基于信息熵的自适应抽样算法。实验结果表明，该方法的效果优于香农信息熵等经典方法。

Adaptive Sampling for Monte Carlo Global Illumination Based on Entropy

Monte Carlo is a good choice to compute the problem of global illumination in the field of realistic image synthesis. However, the image produced by a Monte Carlo-based global illumination algorithm is noisy when not using a large enough number of samples. Adaptive sampling is an attractive means to reduce these noises. This paper introduces a new measure based on entropy for adaptive sampling. Experimental results show that the method can perform better than classic ones.