Enhanced anti-aliasing by varying sample patterns spatially and/or temporally

Patent metadata

Field	Value
Publication number	US-10957078-B2
Application number	US-201816209895-A
Country	US
Kind code	B2
Filing date	Dec 4, 2018
Priority date	Sep 10, 2014
Publication date	Mar 23, 2021
Grant date	Mar 23, 2021

Abstract

A raster unit is configured to generate different sample patterns for adjacent pixels within a given frame. In addition, the raster unit may adjust the sample patterns between frames. The raster unit includes an index unit that selects a sample pattern table for use with a current frame. For a given pixel, the index unit extracts a sample pattern from the selected sample pattern table. The extracted sample pattern is used to generate coverage information for the pixel. The coverage information for all pixels is then used to generate an image. The resultant image may then be filtered to reduce or remove artifacts induced by the changing of sample locations.

First claim

What is claimed is:

1. A graphics system, comprising:
a rasterizer that is configured to generate samples for a first image frame based on one or more sample patterns to produce a first multi-sampled image frame and to generate samples for a second image frame based on one or more sample patterns to produce a second multi-sampled image frame, the second image frame being different from the first image frame; and
a device driver that is configured to:
compare the first multi-sampled image frame to the second multi-sampled image frame to detect an amount of motion experienced by a first pixel between the first multi-sampled image frame and the second multi-sampled image frame;
and

based on the amount of motion detected, cause at least one of a first filtering operation or a second filtering operation to be performed, wherein the first filtering operation involves the first pixel and at least one other pixel residing in the second multi-sampled image frame, and the second filtering operation involves the first pixel and at least one other pixel residing in the first multi-sampled image frame.

2. The system of claim 1 , wherein each of the first filtering operation and the second filtering operation, when performed, reduces at least one artifact appearing between the first multi-sampled image frame and the second multi-sampled image frame.

3. The system of claim 1 , wherein the device driver is configured to cause at least one of the first filtering operation or the second filtering operation to be performed by:
determining that the amount of motion does not exceed a threshold level, and causing the first filtering operation to be performed; or
determining that the amount of motion exceeds the threshold level, and causing the second filtering operation to be performed.

4. The system of claim 1 , wherein the device driver is configured to compare the first multi-sampled image frame to the second multi-sampled image frame by:
comparing a first value associated with the first pixel to a second value associated with a second pixel, wherein the second pixel resides in the second multi-sampled image frame; and
determining a difference between the first value and the second value.

5. The system of claim 4 , wherein the first pixel resides at a first position within the first multi-sampled image frame, and the second pixel resides at the first position within the second multi-sampled image frame.

6. The system of claim 1 , wherein the first filtering operation is performed by:
identifying a previous pixel residing in a previous multi-sampled image frame generated before the first multi-sampled image frame; and
blending the first pixel with the previous pixel.

7. The system of claim 1 , wherein the second filtering operation is performed by:
identifying an edge within the first multi-sampled image frame along which the first pixel and one or more other pixels reside; and
blending the first pixel with the one or more other pixels.

8. The system of claim 1 , wherein the rasterizer is configured to generate samples for the first image frame by:
generating a first index into a first sample pattern table that includes a plurality of entries, wherein the first index references a first subset of entries included in the plurality of entries;
extracting the first subset of entries from the first sample pattern table, wherein a first entry included in the first subset of entries indicates a first location within the first pixel where a first sample is to be generated; and
generating the first sample at the first location.

9. The system of claim 8 , wherein the first sample comprises a coverage sample, and further comprising a shader unit that is configured to generate a shader sample at the first location and to process the shader sample to generate a color value for the first pixel.

10. The system of claim 8 , wherein the rasterizer is further configured to determine a frame number associated with the first image frame and to select the first sample pattern table from a plurality of sample pattern tables based on the first frame number.

11. The system of claim 1 , further comprising a streaming multiprocessor that is configured to perform that at least one of the first filtering operation or the second filtering operation.

12. A method, comprising:
generating samples for a first image frame based on one or more sample patterns to produce a first multi-sampled image frame;
generating samples for a second image frame based on one or more sample patterns to produce a second multi-sampled image frame, the second image frame being different from the first image frame;
comparing the first multi-sampled image frame to the second multi-sampled image frame to detect an amount of motion experienced by a first pixel between the first multi-sampled image frame and the second multi-sampled image frame; and
based on the amount of motion detected, causing at least one of a first filtering operation or a second filtering operation to be performed, wherein the first filtering operation involves the first pixel and at least one other pixel residing in the second multi-sampled image frame, and the second filtering operation involves the first pixel and at least one other pixel residing in the first multi-sampled image frame.

13. The method of claim 12 , wherein each of the first filtering operation and the second filtering operation, when performed, reduces at least one artifact appearing between the first multi-sampled image frame and the second multi-sampled image frame.

14. The method of claim 12 , wherein causing at least one of the first filtering operation or the second filtering operation to be performed comprises:
determining that the amount of motion does not exceed a threshold level, and causing the first filtering operation to be performed; or
determining that the amount of motion exceeds the threshold level, and causing the second filtering operation to be performed.

15. The method of claim 12 , wherein comparing the first multi-sampled image frame to the second multi-sampled image frame comprises:
comparing a first value associated with the first pixel to a second value associated with a second pixel, wherein the second pixel resides in the second multi-sampled image frame; and
determining a difference between the first value and the second value.

16. The method of claim 15 , wherein the first pixel resides at a first position within the first multi-sampled image frame, and the second pixel resides at the first position within the second multi-sampled image frame.

17. The method of claim 12 , further comprising performing the first filtering operation by:
identifying a previous pixel residing in a previous multi-sampled image frame generated before the first multi-sampled image frame; and
blending the first pixel with the previous pixel.

18. The method of claim 12 , further comprising performing the second filtering operation by:
identifying an edge within the first multi-sampled image frame along which the first pixel and one or more other pixels reside; and
blending the first pixel with the one or more other pixels.

19. The method of claim 12 , wherein generating samples for the first image frame comprises:
generating a first index into a first sample pattern table that includes a plurality of entries, wherein the first index references a first subset of entries included in the plurality of entries;
extracting the first subset of entries from the first sample pattern table, wherein a first entry included in the first su

Assignees

• Nvidia Corp

Inventors

• Uralsky Yury Y
• Alben Jonah M
• Banerjee Ankan
• Massal Gregory
• Petersen Thomas
• Kuznetsov Oleg
• Lum Eric B
• Mehta Prakshep

Classifications

• G06T11/10 Primary

  Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title

• G06T11/40 Primary

  Filling planar surfaces by adding surface attributes, e.g. adding colours or textures · CPC title

• G06T2200/12

  involving antialiasing · CPC title

• G06T11/001 Primary

  Physics · mapped topic