Context-aware compression with quantization of hierarchical transform matrices

Patent metadata

Field	Value
Publication number	US-11915459-B2
Application number	US-202217740754-A
Country	US
Kind code	B2
Filing date	May 10, 2022
Priority date	Dec 28, 2018
Publication date	Feb 27, 2024
Grant date	Feb 27, 2024

Abstract

Apparatus and method for context-aware compression. For example, one embodiment of an apparatus comprises: ray traversal/intersection circuitry to traverse rays through a hierarchical acceleration data structure to identify intersections between rays and primitives of a graphics scene; matrix compression circuitry/logic to compress hierarchical transformation matrices to generate compressed hierarchical transformation matrices by quantizing N-bit floating point data elements associated with child transforms of the hierarchical transformation matrices to variable-bit floating point numbers or integers comprising offsets from a parent transform of the child transform; and an instance processor to generate a plurality of instances of one or more base geometric objects in accordance with the compressed hierarchical transformation matrices.

First claim

What is claimed is:

1. An apparatus comprising:
a storage to store one or more base geometric objects of a graphics scene; and
a processor coupled to the storage, the processor to compress initial transformation matrices to generate compressed transformation matrices, wherein the initial transformation matrices are hierarchical transformation matrices, and to generate a plurality of instances of the one or more base geometric objects based on the compressed transformation matrices, wherein the plurality of instances of the one or more base geometric objects are to be stored hierarchically and used to render the graphics scene, and wherein one or more variable bit floating point numbers or integers corresponding to a child transform of the hierarchical transformation matrices comprises one or more offsets from a parent transform of the child transform.

2. The apparatus of claim 1 , wherein compressing the initial transformation matrices comprises reducing precision of elements of the initial transformation matrices based on a required precision.

3. The apparatus of claim 2 , wherein the required precision is based on contextual parameters determined when the graphics scene is created.

4. The apparatus of claim 1 , wherein compressing the initial transformation matrices comprises quantizing fixed-bit length floating point values of the initial transformation matrices to variable bit floating point values or integers.

5. The apparatus of claim 1 , wherein the processor is further to traverse rays through a hierarchical acceleration data structure to identify intersections between the rays and primitives of the graphics scene.

6. The apparatus of claim 5 , wherein current context parameters associated with the ray traversal are analyzed to determine a level of precision to compress the initial transformation matrices to generate the compressed transformation matrices.

7. The apparatus of claim 5 , wherein current context parameters are to be evaluated to determine an error tolerance, the error tolerance based on at least one of a distance from a camera to the primitives of the graphics scene or a type of ray being traversed.

8. The apparatus of claim 1 , wherein compressing the initial transformation matrices comprises procedurally generating the compressed transformation matrices based on the initial transformation matrices.

9. The apparatus of claim 1 , wherein the generated plurality of instances of the one or more base geometric objects are saved in the storage.

10. A method comprising:
storing one or more base geometric objects of a graphics scene to a storage of an apparatus;
compressing initial transformation matrices to generate compressed transformation matrices, wherein the initial transformation matrices are hierarchical transformation matrices, and wherein one or more variable bit floating point numbers or integers corresponding to a child transform of the hierarchical transformation matrices comprises one or more offsets from a parent transform of the child transform;
generating a plurality of instances of the one or more base geometric objects based on the compressed transformation matrices, wherein the plurality of instances of the one or more base geometric objects are to be stored hierarchically; and
rendering the graphics scene based on the plurality of instances of the one or more base geometric objects.

11. The method of claim 10 , wherein compressing the initial transformation matrices comprises reducing precision of elements of the initial transformation matrices based on a required precision.

12. The method of claim 10 , wherein compressing the initial transformation matrices comprises quantizing fixed-bit length floating point values of the initial transformation matrices to variable bit floating point values or integers.

13. The method of claim 10 , further comprising:
traversing rays through a hierarchical acceleration data structure to identify intersections between the rays and primitives of the graphics scene.

14. The method of claim 10 , wherein compressing the initial transformation matrices comprises procedurally generating the compressed transformation matrices based on the initial transformation matrices.

15. The method of claim 10 , wherein the generated plurality of instances of the one or more base geometric objects are saved in the storage.

16. A non-transitory machine-readable medium having program code stored thereon which, when executed by a machine, causes the machine to perform:
storing one or more base geometric objects of a graphics scene to a storage of an apparatus;
compressing initial transformation matrices to generate compressed transformation matrices, wherein the initial transformation matrices are hierarchical transformation matrices, and wherein one or more variable bit floating point numbers or integers corresponding to a child transform of the hierarchical transformation matrices comprises one or more offsets from a parent transform of the child transform;
generating a plurality of instances of the one or more base geometric objects based on the compressed transformation matrices, wherein the plurality of instances of the one or more base geometric objects are to be stored hierarchically; and
rendering the graphics scene based on the plurality of instances of the one or more base geometric objects.

17. The non-transitory machine-readable medium of claim 16 , wherein compressing the initial transformation matrices comprises reducing precision of elements of the initial transformation matrices based on a required precision.

18. The non-transitory machine-readable medium of claim 16 , wherein compressing the initial transformation matrices comprises quantizing fixed-bit length floating point values of the initial transformation matrices to variable bit floating point values or integers.

19. The non-transitory machine-readable medium of claim 16 , wherein the machine is caused to further perform:
traversing rays through a hierarchical acceleration data structure to identify intersections between the rays and primitives of the graphics scene.

Assignees

• Intel Corp

Inventors

• Brownlee Carson
• Benthin Carsten
• Barczak Joshua
• Xiao Kai
• Apodaca Michael
• Surti Prasoonkumar
• Raoux Thomas

Classifications

• G06T9/40 Primary

  Tree coding, e.g. quadtree, octree · CPC title

• G06T1/60 Primary

  Memory management · CPC title

• G06T15/06

  Ray-tracing · CPC title

• G06T15/20

  Perspective computation · CPC title

• G06T2210/21

  Collision detection, intersection · CPC title