VISIBILITY-BASED STATE UPDATES IN GRAPHICAL PROCESSING UNITS

• US 20130241938A1
• Filed: 03/15/2012
• Published: 09/19/2013
• Est. Priority Date: 03/15/2012
• Status: Active Grant

• Current Assignee: Qualcomm, Inc.
• Original Assignee: Qualcomm, Inc.
• Inventors: Gruber, Andrew, Jenkal, Ravi S.
• Granted Patent: US 10,242,481 B2
• Time in Patent Office: Days
• US Class Current: 345/501
• CPC Class Codes: G06T 11/40 Filling a planar surface by adding surface attributes, e.g. colour or texture
• CPC Class Codes: G06T 15/005 General purpose rendering architectures

Abstract

In general, techniques are described for visibility-based state updates in graphical processing units (GPUs). A device that renders image data comprising a memory configured to store state data and a GPU may implement the techniques. The GPU may be configured to perform a multi-pass rendering process to render an image from the image data. The GPU determines visibility information for a plurality of objects defined by the image data during a first pass of the multi-pass rendering process. The visibility information indicates whether each of the plurality of objects will be visible in the image rendered from the image data during a second pass of the multi-pass rendering process. The GPU then retrieves the state data from the memory for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data based on the visibility information.

Claims

1. A method for rendering image data comprising:

determining, in a graphics processing unit, visibility information for a plurality of objects defined by the image data during a first pass of a multi-pass rendering process, wherein the visibility information indicates whether each of the plurality of objects will be visible in an image rendered from the image data during a second pass of the multi-pass rendering process; and

retrieving state data, in the graphics processing unit, for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data based on the visibility information, wherein the first pass finishes prior to the completion of the second pass.

2. The method of claim 1, further comprising segmenting the image data into a plurality of tiles,

wherein determining visibility information comprises determining tile-specific visibility information for each of the plurality of tiles during the first pass of the multi-pass rendering process, wherein the tile-specific visibility information indicates whether, for a corresponding one of the plurality of tiles, each of the plurality of objects will be visible in a corresponding tile of the image rendered from the image data during the second pass of the multi-pass rendering process, and

wherein retrieving state data comprises retrieving state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data based on the corresponding tile-specific visibility information to output the respective one of the plurality of tiles of the image.

3. The method of claim 1,

wherein the graphics processing unit implements the multi-pass rendering process,

wherein at least one of the plurality of objects includes a pointer that identifies a starting location in a memory external from the graphics processing unit that stores the state data, and

wherein retrieving the state data comprises dereferencing the pointer to access the starting location in the memory external from the graphics processing unit and retrieve the state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data when the visibility information indicates that a corresponding one of the plurality of objects is visible.

4. The method of claim 1, wherein at least two or more of the plurality of objects share the same state data.

5. The method of claim 1,

wherein the image data defines the plurality of objects in model space,

wherein the method further comprises transforming the plurality of objects from the model space to a screen space so as to generate transformed image data that defines a plurality of transformed objects in the screen space, and

wherein determining the visibility information comprises determining visibility information for the plurality of transformed objects defined by the transformed image data during the first pass of the multi-pass rendering process.

6. The method of claim 1, wherein determining the visibility information comprises determining the visibility information for the plurality of objects defined by the image data prior to rendering the plurality of objects defined by the image data to generate the image.

7. The method of claim 1, wherein determining the visibility information comprises performing a low resolution Z-buffer process to determine the visibility information.

8. The method of claim 1, wherein the GPU resides within a mobile computing device.

9. A device that renders image data, the device comprising:

means for determining visibility information for a plurality of objects defined by the image data during a first pass of a multi-pass rendering process, wherein the visibility information indicates whether each of the plurality of objects will be visible in an image rendered from the image data during a second pass of the multi-pass rendering process; and

means for retrieving state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data based on the visibility information, wherein the first pass finishes prior to the completion of the second pass.

10. The device of claim 9, further comprising means for segmenting the image data into a plurality of tiles,

wherein the means for determining visibility information comprises means for determining tile-specific visibility information for each of the plurality of tiles during the first pass of the multi-pass rendering process, wherein the tile-specific visibility information indicates whether, for a corresponding one of the plurality of tiles, each of the plurality of objects will be visible in a corresponding tile of the image rendered from the image data during the second pass of the multi-pass rendering process, and

wherein the means for retrieving state data comprises means for retrieving state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data based on the corresponding tile-specific visibility information to output the respective one of the plurality of tiles of the image.

11. The device of claim 9,

wherein a graphics processing unit implements the multi-pass rendering process and includes the means for determining the visibility information and the means for retrieving state data,

wherein at least one of the plurality of objects includes a pointer that identifies a starting location in a memory external from the graphics processing unit that stores the state data, and

wherein the means for retrieving the state data comprises means for dereferencing the pointer to access the starting location in the memory external from the graphics processing unit and retrieve the state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data when the visibility information indicates that a corresponding one of the plurality of objects is visible.

12. The device of claim 9, wherein at least two or more of the plurality of objects share the same state data.

13. The device of claim 9,

wherein the image data defines the plurality of objects in model space,

wherein the device further comprises means for transforming the plurality of objects from the model space to a screen space so as to generate transformed image data that defines a plurality of transformed objects in the screen space, and

wherein the means for determining the visibility information comprises means for determining visibility information for the plurality of transformed objects defined by the transformed image data during the first pass of the multi-pass rendering process.

14. The device of claim 9, wherein the means for determining the visibility information comprises means for determining the visibility information for the plurality of objects defined by the image data prior to rendering the plurality of objects defined by the image data to generate the image.

15. The device of claim 9, wherein the means for determining the visibility information comprises means for performing a low resolution Z-buffer process to determine the visibility information.

16. The device of claim 9, wherein the device comprises a mobile computing device.

17. A device that renders image data, the device comprising:

a memory configured to store state data; and

a processor configured to perform a multi-pass rendering process to render an image from the image data, wherein the processor determines visibility information for a plurality of objects defined by the image data during a first pass of the multi-pass rendering process, wherein the visibility information indicates whether each of the plurality of objects will be visible in the image rendered from the image data during a second pass of the multi-pass rendering process and retrieves the state data from the memory for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data based on the visibility information, wherein the first pass finishes prior to the completion of the second pass.

18. The device of claim 17, further comprising segmenting the image data into a plurality of tiles,

wherein the processor is further configured to determine tile-specific visibility information for each of the plurality of tiles during the first pass of the multi-pass rendering process, wherein the tile-specific visibility information indicates whether, for a corresponding one of the plurality of tiles, each of the plurality of objects will be visible in a corresponding tile of the image rendered from the image data during the second pass of the multi-pass rendering process, and

wherein the processor is also configured to retrieve state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data based on the corresponding tile-specific visibility information to output the respective one of the plurality of tiles of the image.

19. The device of claim 17,

wherein the processor comprises a graphics processing unit,

wherein at least one of the plurality of objects includes a pointer that identifies a starting location in a memory external from the graphics processing unit that stores the state data, and

wherein the graphics processing unit is configured to dereference the pointer to access the starting location in the memory external from the graphics processing unit and retrieve the state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data when the visibility information indicates that a corresponding one of the plurality of objects is visible.

20. The device of claim 17, wherein at least two or more of the plurality of objects share the same state data.

21. The device of claim 17,

wherein the image data defines the plurality of objects in model space,

wherein the processor is further configured to transform the plurality of objects from the model space to a screen space so as to generate transformed image data that defines a plurality of transformed objects in the screen space, and determine visibility information for the plurality of transformed objects defined by the transformed image data during the first pass of the multi-pass rendering process.

22. The device of claim 17, wherein the processor is configured to determine the visibility information for the plurality of objects defined by the image data prior to rendering the plurality of objects defined by the image data to generate the image.

23. The device of claim 17, wherein the processor is further configured to perform a low resolution Z-buffer process to determine the visibility information.

24. The device of claim 17, wherein the device comprises a mobile computing device.

25. A non-transitory computer-readable medium comprising instructions that, when executed, cause one or more processors to:

determine visibility information for a plurality of objects defined by the image data during a first pass of a multi-pass rendering process, wherein the visibility information indicates whether each of the plurality of objects will be visible in an image rendered from the image data during a second pass of the multi-pass rendering process; and

retrieve state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data based on the visibility information, wherein the first pass finishes prior to the completion of the second pass.

26. The non-transitory computer-readable medium of claim 25, further comprising instructions that, when executed, cause the one or more processors to:

segment the image data into a plurality of tiles;

determine tile-specific visibility information for each of the plurality of tiles during the first pass of the multi-pass rendering process, wherein the tile-specific visibility information indicates whether, for a corresponding one of the plurality of tiles, each of the plurality of objects will be visible in a corresponding tile of the image rendered from the image data during the second pass of the multi-pass rendering process, and

retrieve state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data based on the corresponding tile-specific visibility information to output the respective one of the plurality of tiles of the image.

27. The non-transitory computer-readable medium of claim 25,

wherein the one or more processors comprises at least one graphics processing units,

wherein at least one of the plurality of objects includes a pointer that identifies a starting location in a memory external from the graphics processing unit that stores the state data, and

wherein the non-transitory computer-readable medium further comprises instructions that, when executed, cause the at least one graphics processing unit to retrieve the state data comprises dereferencing the pointer to access the starting location in the memory external from the graphics processing unit and retrieve the state data for use by the second pass of the multi-pass rendering process in rendering the plurality of objects of the image data when the visibility information indicates that a corresponding one of the plurality of objects is visible.

28. The non-transitory computer-readable medium of claim 25, wherein at least two or more of the plurality of objects share the same state data.

29. The non-transitory computer-readable medium of claim 25,

wherein the image data defines the plurality of objects in model space,

wherein the non-transitory computer-readable medium further comprises instructions that, when executed, cause the one or more processors to:

transform the plurality of objects from the model space to a screen space so as to generate transformed image data that defines a plurality of transformed objects in the screen space; and

determine visibility information for the plurality of transformed objects defined by the transformed image data during the first pass of the multi-pass rendering process.

30. The non-transitory computer-readable medium of claim 25, further comprising instructions that, when executed, cause the one or more processors to determine the visibility information for the plurality of objects defined by the image data prior to rendering the plurality of objects defined by the image data to generate the image.

31. The non-transitory computer-readable medium of claim 25, further comprising instructions that, when executed, cause the one or more processors to perform a low resolution Z-buffer process to determine the visibility information.

32. The non-transitory computer-readable medium of claim 25, wherein a mobile computing device includes the non-transitory computer-readable medium and the one or more processors.