Efficient Differentiation of Pixel Reconstruction Filters for Path-Space Differentiable Rendering

Efficient Differentiation of Pixel Reconstruction Filters for Path-Space Differentiable Rendering
复制标题

用于路径空间可微渲染的像素重建滤波器的高效微分

DOI:
10.1145/3550454.3555500
复制
发表时间:
2022
影响因子:
6.2
通讯作者:
Zhao, Shuang
Zhao, Shuang
中科院分区:
计算机科学1区
文献类型:
--
作者:
Yu, Zihan;Zhang, Cheng;Nowrouzezahrai, Derek;Dong, Zhao;Zhao, Shuang

文献摘要

参考文献

被引文献

相似文献

像素重建滤波器在基于物理的渲染中起着重要的作用,并且已经得到了深入的研究。然而,在基于物理的可微分渲染中,对像素过滤器的适当处理在很大程度上仍未得到充分探索。提出了一种基于路径空间公式的像素重构滤波器的高效差分方法。具体来说,我们制定了像素滤波器中不连续的像素边界积分,并引入了支持可微路径采样方法的新的对偶采样方法,如伴随粒子跟踪和双向路径跟踪。在估计这个积分时,我们证明了反采样的必要性和有效性,并评估了它在几个可微渲染和反渲染设置中的有效性。
Pixel reconstruction filters play an important role in physics-based rendering and have been thoroughly studied. In physics-based differentiable rendering, however, the proper treatment of pixel filters remains largely under-explored. We present a new technique to efficiently differentiate pixel reconstruction filters based on the path-space formulation. Specifically, we formulate the pixelboundaryintegral that models discontinuities in pixel filters and introduce new antithetic sampling methods that support differentiable path sampling methods, such as adjoint particle tracing and bidirectional path tracing. We demonstrate both the need and efficacy of antithetic sampling when estimating this integral, and we evaluate its effectiveness across several differentiable- and inverse-rendering settings.
DOI: --
发表时间: 2014
影响因子: 6.2
作者:
Laurent Belcour;Kavita Bala;C. Soler
通讯作者: C. Soler
DOI: 10.1145/3528223.3530080
发表时间: 2022
影响因子: 6.2
作者:
Yan, Kai;Lassner, Christoph;Budge, Brian;Dong, Zhao;Zhao, Shuang
通讯作者: Zhao, Shuang
参与媒体的路径空间可微渲染
DOI: 10.1145/3450626.3459782
发表时间: 2021
影响因子: 6.2
作者:
Zhang, Cheng;Yu, Zihan;Zhao, Shuang
通讯作者: Zhao, Shuang
尽可能稀疏的 SVBRDF 采集
DOI: --
发表时间: 2016
影响因子: 6.2
作者:
Zhiming Zhou;Guojun Chen;Yue Dong;D. Wipf;Yong Yu;John M. Snyder;Xin Tong
通讯作者: Xin Tong
DOI: 10.1145/2461912.2461943
发表时间: 2013
期刊: ACM Transactions on Graphics (TOG)
影响因子: --
作者:
J. Lehtinen;Tero Karras;S. Laine;M. Aittala;F. Durand;Timo Aila
通讯作者: Timo Aila