Binh-Son Hua (1) Adrien Gruson(1,4) Victor Petitjean (2)
Matthias Zwickerv(3) Derek Nowrouzezahrai (4) Elmar Eisemann (2) Toshiya Hachisuka (1)
1.The University of Tokyo 
2.Delft University of Technology 
3.University of Maryland, College Park 
4.McGill University

Monte Carlo methods for physically-based light transport simulation are broadly adopted in the feature film production, animation and visual effects industries. These methods, however, often result in noisy images（噪声图像） and have slow convergence（收敛慢）. As such, improving the convergence of Monte Carlo rendering remains an important open problem. Gradient-domain light transport（梯度域光线传输） is a recent family of techniques that can accelerate Monte Carlo rendering by up to an order of magnitude（一个数量级）, leveraging a gradient-based estimation（梯度与估计） and a reformulation of the rendering problem as an image reconstruction（渲染问题看作图像重建）. This state of the art report comprehensively frames the fundamentals of gradient-domain rendering（梯度域渲染的基础）, as well as the pragmatic details behind practical gradient-domain uni and bidirectional path tracing（梯度域单向双向路径追踪的实际细节） and photon density estimation（光子密度估计） algorithms. Moreover, we discuss the various image reconstruction schemes（各种图像重建方案） that are crucial to accurate and stable gradient-domain rendering. Finally, we benchmark various gradient-domain techniques against the state-of-the-art in denoising methods（去噪方法） before discussing open problems.

Section 2 梯度域渲染的一般概念和成分
Section 3 梯度估计的基本成分：高效移位映射（effective shift mappings）
Section 4 梯度域光子密度估计和梯度域顶点连接合并
Section 5 各种梯度域图像重建方法
Section 6 基于蒙特克罗图像去噪和高级梯度采样方法的图像重建算法
Section 7 volumetric participating media、时间图像序列temporal image sequences和谱渲染spectral rendering场景下的光线传播
Section 8 总结与展望