# US 2023/0066636 A1 · Inverse transform sampling through ray tracing

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2023/0066636 A1
- Document ID: US20230066636A1
- Status: Published application
- Kind: A1
- Application number: 17/947,907
- Filed: 2022-09-19
- Published: 2023-03-02
- Earliest claimed: 2021-08-26
- Assignees: NVIDIA Corporation
- Applicants: Nvidia Corporation
- Inventors: Nathan Morrical
- Web page: https://patentgrep.com/patent/US20230066636A1

## CPC

- G06T 15/06: Ray-tracing
- G06T 15/04: Texture mapping
- G06T 15/506: {Illumination models}
- G06T 2210/21: Collision detection, intersection
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

High quality image rendering can be achieved in part by using inverse transform sampling to direct sampling toward regions of greater importance, such as regions with higher brightness values, to reduce noise and improve convergence. Inverse transform sampling can be achieved more efficiently by reformulating as a ray-tracing problem, using tree traversal units that can be accelerated. A geometric mesh can be generated based on a set of cumulative distribution functions (CDFs) for various rows and columns of pixels in a texture, and individual rays can be traced against this mesh, with those rays having a higher probability of intersection at a point with greater importance, such as a higher brightness value. A probability distribution function to be used for importance sampling can be derived by analyzing partial derivatives of the CDF geometry at the intersection location.

## Claims

### Claim 1

A computer-implemented method, comprising:
determining importance values for one or more pixels of an image;
generating, based at least in part on the importance values, a geometric mesh of the image corresponding to interaction probabilities for the one or more pixels;
importance sampling the geometric mesh using one or more simulated light transport rays; and
determining, using the importance sampling, one or more pixel locations corresponding to interactions between the one or more simulated light transport rays and the geometric mesh.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, further comprising:
determining, based at least in part, on the importance values, a set of cumulative distribution functions (CDFs) for the one or more pixel locations.

### Claim 3 (depends on 2)

The computer-implemented method of claim 2, wherein the set of CDFs is determined based at least on probability distributions of the importance values for individual rows and columns of the image.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein the importance sampling comprises a ray traversal search with a random number of simulated light transport rays.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, further comprising:
rendering the image using, at least in part, a representation of illumination at the one or more pixel locations associated with the interactions.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the importance values correspond to grayscale luminance values determined for the one or more pixels of the image.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the geometric mesh includes a series of triangulated ribbons corresponding to rows or columns of the pixels.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein at least the determining of the interactions is performed using hardware acceleration.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, wherein the importance sampling comprises Monte Carlo-based importance sampling, the Monte Carlo-based importance sampling comprising a probability distribution function derived based, at least on one or more partial derivatives of the geometric mesh at one or more of the intersection points.

### Claim 10

A system, comprising:
one or more processing units to:
determine importance values for one or more pixels of an image;
generate, based at least in part on the importance values, a geometric mesh of the image corresponding to interaction probabilities for the one or more pixels;
importance sampling the geometric mesh using one or more simulated light transport rays; and
determine, using the importance sampling, one or more pixel locations corresponding to interactions between the one or more simulated light transport rays and the geometric mesh.

### Claim 11 (depends on 10)

The system of claim 10, wherein the system is comprised in at least one of:
a system for performing simulation operations;
a system for performing simulation operations to test or validate autonomous machine applications;
a system for rendering graphical output;
a system for performing deep learning operations;
a system implemented using an edge device;
a system incorporating one or more Virtual Machines (VMs);
a system implemented at least partially in a data center; or
a system implemented at least partially using cloud computing resources.

### Claim 12 (depends on 10)

The system of claim 10, wherein the importance values correspond to grayscale luminance values determined for the respective pixels of the image.

### Claim 13 (depends on 10)

The system of claim 10, wherein the one or more processing units are further to determine, based at least on the importance values, a set of cumulative distribution functions (CDFs) for the one or more pixel locations.

### Claim 14 (depends on 13)

The system of claim 13, wherein the set of CDFs is determined based at on probability distributions of the importance values for individual rows and columns of the image.

### Claim 15 (depends on 10)

The system of claim 10, wherein at least the determining of the interactions is performed using hardware acceleration.

### Claim 16 (depends on 10)

The system of claim 10, wherein the one or more processing units are further to render the image using, at least in part, a representation of illumination at the one or more pixel locations associated with the interactions.

### Claim 17

A processor, comprising:
one or more processing units to determine interactions between one or more simulated light transport rays and one or more pixel locations of a geometric mesh, geometric mesh corresponding to interaction probabilities for the one or more pixel locations based at on importance values corresponding to respective pixel locations of the one or more pixel locations.

### Claim 18 (depends on 17)

The processor of claim 17, wherein the processor is comprised in at least one of:
a system for performing simulation operations;
a system for performing simulation operations to test or validate autonomous machine applications;
a system for rendering graphical output;
a system for performing deep learning operations;
a system implemented using an edge device;
a system incorporating one or more Virtual Machines (VMs);
a system implemented at least partially in a data center; or
a system implemented at least partially using cloud computing resources.

### Claim 19 (depends on 17)

The processor of claim 17, wherein the processor further determines, based at least on the importance values, a set of cumulative distribution functions (CDFs) for the one or more pixel locations.

### Claim 20 (depends on 19)

The processor of claim 19, wherein the set of CDFs is determined based at least on probability distributions of the importance values for individual rows and columns of the image.
