使用神经辐射先验的全局反照率反演（英文版）

英文标题：Inverse Global Illumination using a Neural Radiometric Prior中文摘要：本研究提出了一种基于辐射先验和神经网络的简单的反渲染方法，可正确计算全局光照，同时在更短的计算时间内实现了与传统方法相当甚至更好的场景参数恢复质量。英文摘要：Inverse rendering methods that account for global illumination are becomingmore popular, but current methods require evaluating and automaticallydifferentiating millions of path integrals by tracing multiple light bounces,which remains expensive and prone to noise. Instead, this paper proposes aradiometric prior as a simple alternative to building c