神经求交函数
Neural Intersection Function
AI总结:
本文提出基于多层感知机的神经求交函数,用稠密矩阵乘法替代BVH中最不规则耗时的遍历部分,在统一渲染框架中加速次级光线投射,实验显示直接光照次级光线投射时间最多减少35%且保持图像质量。
摘要:
蒙特卡洛光线追踪算法中的光线投射操作通常采用层次包围体(BVH)来加速求交过程以评估可见性。然而,其特性是不规则的,在内存访问和分支执行上存在发散,因此无法在GPU上达到最高效率。本文提出了一种新颖的基于多层感知机的神经求交函数,其核心运算仅包含具有可预测内存访问模式的稠密矩阵乘法。我们的方法是首个将基于神经网络的方法与基于BVH的光线追踪管线集成到统一渲染框架中的解决方案。我们能够在不遍历BVH中最不规则且最耗时的部分的情况下评估次级光线的可见性和遮挡关系,从而加速光线投射。实验表明,与基于BVH的实现相比,所提方法可将直接光照的次级光线投射时间最多减少35%,同时仍保持图像质量。
英文摘要:
The ray casting operation in the Monte Carlo ray tracing algorithm usually adopts a bounding volume hierarchy (BVH) to accelerate the process of finding intersections to evaluate visibility. However, its characteristics are irregular, with divergence in memory access and branch execution, so it cannot achieve maximum efficiency on GPUs. This paper proposes a novel Neural Intersection Function based on a multilayer perceptron whose core operation contains only dense matrix multiplication with predictable memory access. Our method is the first solution integrating the neural network-based approach and BVH-based ray tracing pipeline into one unified rendering framework. We can evaluate the visibility and occlusion of secondary rays without traversing the most irregular and time-consuming part of the BVH and thus accelerate ray casting. The experiments show the proposed method can reduce the secondary ray casting time for direct illumination by up to 35% compared to a BVH-based implementation and still preserve the image quality.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。