Online Path Sampling Control with Progressive Spatio-temporal Filtering
- PDF / 2,731,046 Bytes
- 16 Pages / 595.276 x 790.866 pts Page_size
- 67 Downloads / 201 Views
ORIGINAL RESEARCH
Online Path Sampling Control with Progressive Spatio‑temporal Filtering Jacopo Pantaleoni1 Received: 15 June 2020 / Accepted: 6 August 2020 © Springer Nature Singapore Pte Ltd 2020
Abstract This work introduces a novel technique called progressive spatio-temporal filtering, an efficient method to build all-frequency approximations to the light transport distribution into a scene by filtering individual samples produced by an underlying path sampler, using online, iterative algorithms and data-structures that exploit both the spatial and temporal coherence of the approximated light field. Unlike previous approaches, the proposed method is both more efficient, due to its use of an iterative temporal feedback loop that massively improves convergence to a noise-free approximant, and more flexible, due to its introduction of a spatio-directional hashing representation that allows to encode directional variations like those due to glossy reflections. We then introduce four different methods to employ the resulting approximations to control the underlying path sampler and/or modify its associated estimator, greatly reducing its variance and enhancing its robustness to complex lighting scenarios. The core algorithms are highly scalable and low-overhead, requiring only minor modifications to an existing path tracer. Combined with the most recent advances in ray-tracing hardware and denoising algorithms, our work enables performing high-quality light transport simulations including complex visibility and transport phenomena in real time. Keywords Global illumination · Light transport simulation · Markov Chain Monte Carlo
Introduction Light transport simulation can be an arbitrarily challenging problem, due to the fact it requires to numerically estimate millions of pixel integrals whose infinite dimensional integrands may have arbitrarily high variance. Forty years of research have produced a vast plethora of methods to increase the efficiency of this complex estimation problem, mostly based on variants of Monte Carlo integration methods, often tailored to specific scenarios. The vast majority of these propose different strategies for path sampling, the core operation required to numerically sample the pixel integrals. In this category fall many general purpose methods, like bidirectional path tracing and its variants [30], MCMC techniques like Metropolis Light Transport and its descendents [4, 20, 27, 31], as well as more ad hoc methods such as many-lights sampling, manifold exploration [12, 16, 18], and many others.
* Jacopo Pantaleoni [email protected] 1
Despite the sheer amount of research, the most popular basic method for path sampling remains path tracing [17], often augmented by specific techniques to sample particular light transport events. The reason why the most basic technique is also the most successful is to be found both in its simplicity, which leads to higher execution efficiency on modern computing architectures, and to its very high persample efficiency on average content,
Data Loading...
