2022
DOI: 10.1111/cgf.14582
|View full text |Cite
|
Sign up to set email alerts
|

Path Guiding with Vertex Triplet Distributions

Abstract: Good importance sampling strategies are decisive for the quality and robustness of photorealistic image synthesis with Monte Carlo integration. Path guiding approaches use transport paths sampled by an existing base sampler to build and refine a guiding distribution. This distribution then guides subsequent paths in regions that are otherwise hard to sample. We observe that all terms in the measurement contribution function sampled during path construction depend on at most three consecutive path vertices. We … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
3
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
5

Relationship

2
3

Authors

Journals

citations
Cited by 6 publications
(3 citation statements)
references
References 32 publications
(56 reference statements)
0
3
0
Order By: Relevance
“…Various approaches have been presented, which differ in used spatial or directional data structures [Jen95; LW95; HP02; BDC12; MGN17], domains [ZZ18; RHJD18], or target functions [RGH*20]. Closely related to this work are approaches based on parametric mixture models [VKŠ*14; HEV*16; HZE*19; RHL20; DPÖM21; SHJD22] of distributions from the exponential family, like Gaussian or von‐Mises Fisher (vMF). They are usually fitted using weighted expectation maximization (wEM) [VKŠ*14].…”
Section: Previous Workmentioning
confidence: 99%
“…Various approaches have been presented, which differ in used spatial or directional data structures [Jen95; LW95; HP02; BDC12; MGN17], domains [ZZ18; RHJD18], or target functions [RGH*20]. Closely related to this work are approaches based on parametric mixture models [VKŠ*14; HEV*16; HZE*19; RHL20; DPÖM21; SHJD22] of distributions from the exponential family, like Gaussian or von‐Mises Fisher (vMF). They are usually fitted using weighted expectation maximization (wEM) [VKŠ*14].…”
Section: Previous Workmentioning
confidence: 99%
“…Early works guided new samples based on stored rays or photon maps [LW95, Jen95]. The radiance can be better modelled by the Gaussian Mixture Model (GMM) [VKv*14, SHJD22], or by a spatial‐directional tree [MGN17]. Reibold et al.…”
Section: Related Workmentioning
confidence: 99%
“…Some approaches further take consideration of the correlation between consecutive vertices in the full path, e.g., using product importance sampling [Herholz et al 2016], parallax-aware warping [Ruppert et al 2020] and spatial correlation [Dodik et al 2021;Schüssler et al 2022]. Recently, utilized representative specular paths to enable effectively guided rendering of caustics,…”
Section: Related Workmentioning
confidence: 99%