Outer Rim Archives
Archives · 2017 · 9665974

Granted patent

Methods and systems of joint path importance sampling

Number
9665974
Published
2017-05-30
Filed
2014-05-20
Assignee
DISNEY ENTERPRISES, INC.
Inventors
Jarosz; Wojciech, Georgiev; Iliyan, Krivanek; Jaroslav, Hachisuka; Toshiya, Nowrouzezahrai; Derek
CPC
G06T15/506
Verdict
Low Notable software
Source
Google Patents · FreePatentsOnline

The keeper's note

A joint path importance sampling rendering method for constructing light paths through participating media like fog or smoke.

Abstract

Methods and systems of joint path importance sampling are provided to construct light paths in participating media. The product of anisotropic phase functions and geometric terms across a sequence of path vertices are considered. A connection subpath is determined to join a light source subpath with a light receiver subpath with multiple intermediate vertices while considering the product of phase functions and geometry terms. A joint probability density function (“PDF”) may be factorized unidirectional or bidirectional. The joint PDF may be factorized into a product of multiple conditional PDFs, each of which corresponds to a sampling routine. Analytic importance sampling may be performed for isotropic scattering, whereas tabulated importance sampling may be performed for anisotropic scattering.

Background

BRIEF DESCRIPTION OF THE DRAWINGS(1) The present application, in accordance with one or more various embodiments, is described in detail with reference to the following figures. The drawings are provided for purposes of illustration only and merely depict typical or example embodiments of the application. These drawings are provided to facilitate the reader's understanding of the application and shallnot be considered limiting of the breadth, scope, or applicability of the application. Itshould be noted that for clarity and ease of illustration these drawings are not necessarily made to scale.(2) FIG. 1 illustrates an exemplary configuration for joint path importance sampling.(3) FIG. 2 is a flow diagram illustrating an exemplary method 100 of joint path importance sampling.(4) FIG. 3A illustrates an exemplary process of unidirectional factorization of the joint PDF.(5) FIG. 3B illustrates that an exemplary geometric configuration for sampling a distance for unidirectional factorization of the joint PDF for sampling the distance t.sub.dc 313.(6) FIG. 3C illustrates that an exemplary geometric configuration for sampling a direction for unidirectional factorization of the joint PDF for sampling the direction ω.sub.cb 314.(7) FIG. 3D illustrates that an exemplary geometric configuration for sampling a distance for unidirectional factorization of the joint PDF for sampling the distance t.sub.cb 312.(8) FIG. 4 illustrates an exemplary process of bidirectional factorization of th

Claims

1. A computer-implemented method of sampling light paths connecting a light source subpath to a light receiver subpath, comprising: receiving an input configuration, the input configuration comprises a first endpoint, a first direction, a second endpoint, and a second direction; sampling a set of vertices to determine a set of subpath segments according to a sampling function defined as a product of a set of geometry terms and a set of phase functions; creating asubpath comprising the set of subpath segments to connect the light source subpath with the light receiver subpath; wherein the sampling function comprises a joint probability density function, and the step of sampling a set of vertices comprises factorizing the joint probability density function into a product of a set of conditional probability density functions; and each of the conditional probability density function corresponds to a distance sampling and a direction sampling. 12. A system of sampling light paths connecting a light source subpath to a light receiver subpath, comprising: non-transitory memory storing aset of instructions; and a processor coupled to the non-transitory memory, wherein the set of instructions are configured to cause the processor to perform: receiving an input configuration, the input configuration comprises a first endpoint, a first direction, a second endpoint, and a second direction; sampling a set of vertices to determine a set of subpath segments according to a sampling function defined as a product of a set of geometry terms and a set of phase functions; creating a subpath comprising the set of subpath segmentsto connect the light source subpath with the light receiver subpath; wherein the samplingfunction comprises a joint probability density function, and the set of instructions are configured to cause the processor to factorize the joint probability density function intoa product of a set of conditional probability density functions when performing the step of sampling a set of vertices.