# US 2021/0241502 A1 · Method for improved handling of texture data for texturing and other image processing tasks

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

- Number: US 2021/0241502 A1
- Document ID: US20210241502A1
- Status: Published application
- Kind: A1
- Application number: 17/163,325
- Filed: 2021-01-29
- Published: 2021-08-05
- Earliest claimed: 2020-01-30
- Assignees: Weta Digital Limited
- Applicants: Weta Digital Limited
- Inventors: Kimball D. Thurston, III; Luca Fascione; Sébastien Nicolas Speierer
- Web page: https://patentgrep.com/patent/US20210241502A1

## CPC

- G06T 11/10: Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77)
- G06T 7/40: Analysis of texture (depth or shape recovery from texture G06T7/529)
- G06T 15/04: Texture mapping
- G06T 2210/36: Level of detail
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

An imagery processing system that combines MIP level filtering with spatial filtering when rendering images. Filtering can be performed in an order that optimizes memory accesses during the rendering process.

## Claims

### Claim 1

A computer-implemented method of processing textures to determine a sample, the method comprising:
under the control of one or more computer systems configured with executable instructions:
receive a filter region representation, wherein the filter region representation comprises computer-readable data corresponding to sample region of a virtual mapping to be sampled;
determining a first texture dataset;
determining a filter kernel for the filter region representation, the filter kernel corresponding to the sample region, wherein the filter kernel comprises one or more kernel weights over the sample region;
storing the filter kernel;
determining a subset of the first texture dataset upon which the filter kernel is to be applied;
storing a schedule of the subset, wherein positions of texels within the first texture dataset are determinable from the schedule;
applying the schedule and the filter kernel to the first texture dataset to determine a first texture contribution for the sample region;
determining a second texture dataset;
applying the schedule and the filter kernel to the second texture dataset to determine a second texture contribution for the sample region; and
combining the first texture contribution and the second texture contribution to determine an overall accumulated output for the sample region as the sample.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the texture dataset comprises a plurality of levels of varying resolution and the schedule indicates which level or levels apply to the sample region.

### Claim 3 (depends on 2)

The computer-implemented method of claim 2, wherein the plurality of levels of varying resolution comprises a MIP-map structure, a RIP-map structure, and/or a wavelet structure.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, wherein the first texture dataset and/or the second texture dataset comprises a two-dimensional or a three-dimensional texture.

### Claim 5 (depends on 1)

The computer-implemented method of claim 1, wherein a first subset of the first texture dataset and/or a second subset of the second texture dataset comprises one or more texture tiles of a predetermined size.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the sample region is computed based on a mapping of a region a surface in a virtual scene represented by geometrical models.

### Claim 7 (depends on 1)

The computer-implemented method of claim 1, wherein the filter is determined based on a determined impact of applying the filter across a range of tile vertices.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, further comprising:
accessing one or more texture tiles of one or more levels of varying resolution determined to correspond to a pixel region selected for filtering for the sample region; and
applying at least one fitted filter to the accessed one or more tiles.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, wherein the sample region is an ellipse.

### Claim 10 (depends on 1)

The computer-implemented method of claim 1, wherein a fitted filter is a first fitted filter and a plurality of fitted filters are calculated and accumulated into a representation of the fitted filter in a cache and accumulating the one or more tiles of one or more levels of varying resolution to access.

### Claim 11 (depends on 10)

The computer-implemented method of claim 10, further comprising:
accumulating a list of the one or more tiles to access for subsequent access of those one or more tiles.

### Claim 12 (depends on 1)

The computer-implemented method of claim 1, further comprising generating a plurality of samples for a plurality of requests for samples and caching data values across requests.

### Claim 13 (depends on 12)

The computer-implemented method of claim 12, further comprising grouping texture dataset requests across the requests.