IP Library › Granted Patent US 11,455,770
Granted Patent B2
US 11,455,770 · App. 17/126,721 · Granted Sep 27, 2022

Method and apparatus for encoding based on shading rates

Inventor: Yejun Guo (Shanghai, CN)
Assignee: INTEL CORPORATION
G06T15/80G06T1/20G06T1/60G06T9/001G06T15/005G09G5/39
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,455,770
App. No.
17/126,721
Granted
Sep 27, 2022
Kind
B2
Abstract

Embodiments are generally directed to methods and apparatuses for encoding based on shading rates. An embodiment of a computing system comprises: a memory; a graphics processing unit (GPU) coupled to the memory, the GPU to render a scene of a graphics application into a color buffer within a frame buffer of the memory; and an encoder coupled to the memory, the encoder to encode the content of the color buffer into a video bitstream based on a plurality of shading rates each corresponding to a pixel in the color buffer.

Claims (38)

1. A computing system, comprising:

a memory;

a graphics processing unit (GPU) coupled to the memory, the GPU to render a scene of a graphics application into a color buffer within a frame buffer of the memory; and

an encoder coupled to the memory, the encoder to encode the content of the color buffer into a video bitstream based on a plurality of shading rates each corresponding to a pixel in the color buffer, wherein the plurality of shading rates are passed to a pixel shader associated with the GPU and written by the pixel shader into a shading rate buffer within the frame buffer of the memory.

2. The computing system of claim 1 , wherein the plurality of shading rates being specified by a screen-space image provided by the graphics application or generated based on a rendered frame for the graphics application, wherein the plurality of shading rates being specified by a draw call from the graphics application, wherein the plurality of shading rates being specified by a vertex shader or a geometry shader in the GPU.

3. The computing system of claim 1 , wherein the plurality of shading rates being determined based on a combination operation of at least two of:

(a) a first set of shading rates specified by a screen-space image provided by the graphics application or generated based on a rendered frame for the graphics application;

(b) a second set of shading rates specified by a draw call from the graphics application; and

(c) a third set of shading rates specified by a vertex shader or a geometry shader in the GPU, wherein the combination operation being selected from a minimum operation, a maximum operation, a pass through operation, an override operation and a sum operation.

4. The computing system of claim 1 , wherein pixels in the color buffer that correspond to shading rates finer than a threshold forming interested regions, wherein each of the plurality of shading rates being mapped to a quantization parameter offset for encoding.

5. A method for data processing, comprising:

rendering a scene of a graphics application into a color buffer within a frame buffer of a memory coupled to a graphics processing unit (GPU); and

encoding the content of the color buffer into a video bitstream based on a plurality of shading rates each corresponding to a pixel in the color buffer, wherein the plurality of shading rates are passed to a pixel shader associated with the GPU and written by the pixel shader into a shading rate buffer within the frame buffer of the memory.

6. The method of claim 5 , wherein the plurality of shading rates being specified by a screen-space image provided by the graphics application or generated based on a rendered frame for the graphics application, wherein the plurality of shading rates being specified by a draw call from the graphics application, wherein the plurality of shading rates being specified by a vertex shader or a geometry shader in the GPU.

7. The method of claim 5 , wherein the plurality of shading rates being determined based on a combination operation of at least two of:

(a) a first set of shading rates specified by a screen-space image provided by the graphics application or generated based on a rendered frame for the graphics application;

(b) a second set of shading rates specified by a draw call from the graphics application; and

(c) a third set of shading rates specified by a vertex shader or a geometry shader in the GPU, wherein the combination operation being selected from a minimum operation, a maximum operation, a pass through operation, an override operation and a sum operation.

8. The method of claim 5 , wherein pixels in the color buffer that correspond to shading rates finer than a threshold forming interested regions, wherein each of the plurality of shading rates being mapped to a quantization parameter offset for encoding.

9. An apparatus comprising:

a memory;

a graphics processing unit (GPU) coupled to the memory, the GPU to render a scene of a graphics application into a color buffer within a frame buffer of the memory; and

an encoder coupled to the memory, the encoder to encode the content of the color buffer into a video bitstream based on a plurality of shading rates each corresponding to a pixel in the color buffer, wherein the plurality of shading rates are passed to a pixel shader associated with the GPU and written by the pixel shader into a shading rate buffer within the frame buffer of the memory.

10. The apparatus of claim 9 , wherein the plurality of shading rates being specified by a screen-space image provided by the graphics application or generated based on a rendered frame for the graphics application, wherein the plurality of shading rates being specified by a draw call from the graphics application, wherein the plurality of shading rates being specified by a vertex shader or a geometry shader in the GPU.

11. The apparatus of claim 9 , wherein the plurality of shading rates being determined based on a combination operation of at least two of:

(a) a first set of shading rates specified by a screen-space image provided by the graphics application or generated based on a rendered frame for the graphics application;

(b) a second set of shading rates specified by a draw call from the graphics application; and

(c) a third set of shading rates specified by a vertex shader or a geometry shader in the GPU, wherein the combination operation being selected from a minimum operation, a maximum operation, a pass through operation, an override operation and a sum operation.

12. The apparatus of claim 9 , wherein pixels in the color buffer that correspond to shading rates finer than a threshold forming interested regions, wherein each of the plurality of shading rates being mapped to a quantization parameter offset for encoding.

13. A non-transitory computer-readable medium stored thereon instructions which, when executed, cause a computing device to perform operations comprising:

rendering a scene of a graphics application into a color buffer within a frame buffer of a memory coupled to a graphics processing unit (GPU); and

encoding the content of the color buffer into a video bitstream based on a plurality of shading rates each corresponding to a pixel in the color buffer, wherein the plurality of shading rates are passed to a pixel shader associated with the GPU and written by the pixel shader into a shading rate buffer within the frame buffer of the memory.

14. The non-transitory computer-readable medium of claim 13 , wherein the plurality of shading rates being specified by a screen-space image provided by the graphics application or generated based on a rendered frame for the graphics application, wherein the plurality of shading rates being specified by a draw call from the graphics application, wherein the plurality of shading rates being specified by a vertex shader or a geometry shader in the GPU.

15. The non-transitory computer-readable medium of claim 13 , wherein the plurality of shading rates being determined based on a combination operation of at least two of:

(a) a first set of shading rates specified by a screen-space image provided by the graphics application or generated based on a rendered frame for the graphics application;

(b) a second set of shading rates specified by a draw call from the graphics application; and

(c) a third set of shading rates specified by a vertex shader or a geometry shader in the GPU, wherein the combination operation being selected from a minimum operation, a maximum operation, a pass through operation, an override operation and a sum operation.

16. The non-transitory computer-readable medium of claim 13 , wherein pixels in the color buffer that correspond to shading rates finer than a threshold forming interested regions, wherein each of the plurality of shading rates being mapped to a quantization parameter offset for encoding.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 18, 2020
From: GUO, YEJUN
To: INTEL CORPORATION
Reel/Frame 054693/0625 →
Priority Claims (1)
CN 202010585840.4 · Jun 24, 2020 · national
Continuity (1)
Related Publication 20210407183A1 · Dec 30, 2021
Cited By (2)
US 12,488,228 US 12,499,605