Granted Patent B2

US 12,439,083 · App. 18/328,954 · Granted Oct 7, 2025

Inventors: Alexandros Tourapis (Los Gatos, CA); Jungsun Kim (San Jose, CA); Khaled Mammou (Danville, CA); David W. Singer (San Francisco, CA); Fabrice A. Robinet (Sunnyvale, CA)

Assignee: Apple Inc.

H04N19/597G06T7/10H04N19/20H04N19/467H04N21/8193

Abstract

Claims (101)

1. A non-transitory computer-readable medium storing program instructions, that when executed on or across one or more processors, cause the one or more processors to:

compress a point cloud, wherein to compress the point cloud, the program instructions cause the one or more processors to, for respective ones of a plurality of respective sets of points of the point cloud:

generate, for the respective ones of the respective sets of points, respective sets of attribute patch images and respective sets of depth patch images comprising attribute information or geometry information for the respective sets of points projected onto respective patch planes;

pack the generated sets of attribute patch images and the generated sets of depth patch images in one or more 2D image frames;

video encode the one or more 2D image frames; and

encode one or more supplementary messages comprising one or more entries, wherein the one or more entries define a particular object of a plurality of objects included in the point cloud and comprises respective patch identifiers for respective patches that correspond to portions of the one or more 2D image frames that comprise a sub-set of the set of packed attribute patch images that correspond to the particular object and a sub-set of the set of packed depth patch images that correspond to the particular object in the point cloud being encoded, wherein the sub-set of packed attribute patch images is a sub-set that includes less than a total number of packed attribute patch images that correspond to the plurality of objects, and wherein the sub-set of the set of packed depth patch images that correspond to the particular object is a sub-set that includes less than a total number of packed depth patch images that correspond to the plurality of objects.

2. The non-transitory computer-readable medium of claim 1 , wherein the one or more entries of the one or more supplementary messages comprise:

a first entry identifying the sub-set of the packed attribute and the sub-set of the packed depth patch images of the one or more 2D image frames that correspond to the particular object in the point cloud being encoded;

a second entry identifying respective portions of an occupancy map that correspond to the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images that correspond to the particular object, wherein the occupancy map indicates portions of the one or more 2D image frames that are packed with patches and portions that are padded; and

a third entry comprising information defining a 3D volume space of the point cloud occupied by the particular object.

3. The non-transitory computer-readable medium of claim 2 , wherein the program instructions, when executed on or across the one or more processors, cause the one or more processors to:

encode an additional supplementary message canceling or updating which ones of the attribute patch images and depth patch images of the one or more 2D image frames are included in the respective sub-sets that correspond to the particular object;

encode an additional supplementary message canceling or updating respective portions of the occupancy map that correspond to attribute and depth patch images of the one or more 2D image frames that correspond to the particular object; or

encode an additional supplementary message canceling or updating a 3D volume space of the point cloud occupied by the particular object.

4. The non-transitory computer-readable medium of claim 2 , wherein the one or more supplementary messages further comprise:

an additional entry comprising an object ID for the particular object,

wherein the first entry identifying the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images of the one or more 2D image frames that correspond to the particular object is linked to the object ID,

wherein the second entry identifying the portions of the occupancy map that correspond to the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images that correspond to the particular object is linked to the object ID; and

wherein the third entry comprising the information defining the 3D volume space occupied by the particular object is linked to the object ID.

5. The non-transitory computer-readable medium of claim 4 , wherein the program instructions, when executed one or across the one or more processors, cause the one or more processors to:

determine an updated characteristic for the particular object linked to the object ID; and

encode an update supplementary message that includes the object ID and information indicating the updated characteristic of the particular object without repeating the signaling of information previously signaled for the particular object linked to the object ID that is still applicable to the particular object.

6. The non-transitory computer-readable medium of claim 4 , wherein the program instructions, when executed one or across the one or more processors, further cause the one or more processors to:

determine a location of the particular object in the point cloud at a subsequent moment in time; and

encode an additional supplementary message comprising information defining a 3D volume space of the point cloud occupied by the particular object at the subsequent moment in time,

wherein the additional supplementary message comprises the object ID for the particular object and does not repeat the first entry and the second entries previously encoded for the particular object in the supplementary message.

7. The non-transitory computer-readable medium of claim 4 , wherein the program instructions, when executed on or across the one or more processors, cause the one or more processors to:

encode separate supplementary messages for:

the first or second entries comprising the information identifying the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images that correspond to the particular object or the information identifying the portions of the occupancy map that correspond to the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images that correspond to the particular object; and

the third entry comprising the information defining the 3D volume space occupied by the particular object.

8. The non-transitory computer-readable medium of claim 4 , wherein the program instructions, when executed on or across the one or more processors, cause the one or more processors to:

encode one or more other supplementary messages, wherein the one or more other supplementary messages identify the object ID and comprise:

an entry indicating a label for the particular object;

an entry indicating a collision shape for the particular object;

an entry indicating whether the particular object is visible or hidden;

an entry indicating a rendering priority for the particular object;

an entry indicating a scaling factor for the 3D volume space; or

an entry indicating a relationship between the particular object and other objects of the plurality of objects in the point cloud or other portions of the point cloud.

9. A device, comprising:

a memory storing program instructions for compressing a point cloud; and

one or more processors,

wherein the program instructions, when executed on or across the one or more processors, cause the one or more processors to:

generate a set of attribute patch images and a set of depth patch images comprising attribute information or geometry information for points of the point cloud projected onto respective patch planes;

pack the generated set of attribute patch images and the generated set of depth patch images in one or more 2D image frames;

video encode the one or more 2D image frames; and

encode one or more supplementary messages comprising one or more entries, wherein the one or more entries define a particular object of a plurality of objects included in the point cloud and comprises respective patch identifiers for respective patches that correspond to portions of the one or more 2D image frames that comprise a sub-set of the packed attribute patch images that correspond to the particular object and a sub-set of the packed depth patch images that correspond to the particular object in the point cloud, wherein the sub-set of packed attribute patch images is a sub-set that includes less than a total number of packed attribute patch images that correspond to the plurality of objects, and wherein the sub-set of the set of packed depth patch images that correspond to the particular object is a sub-set that includes less than a total number of packed depth patch images that correspond to the plurality of objects.

10. The device of claim 9 , wherein the program instructions, when executed on or across the one or more processors, cause the one or more processors to:

encode in the one or more supplementary messages, or in one or more update supplementary messages, one or more entries comprising:

an object ID for the particular object;

an object label for the particular object;

a 3D volume space in the point cloud corresponding to the particular object;

a flag indicating whether the particular object is visible or hidden;

a rendering priority for the particular object;

a dependency flag, indicating that rendering the particular object is dependent upon another object being rendered;

a collision shape for the particular object; or

a material ID for the particular object.

11. The device of claim 9 , wherein the one or more entries of the one or more supplementary messages comprise:

a first entry identifying the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images of the one or more 2D image frames that correspond to the particular object in the point cloud being encoded;

a second entry identifying respective portions of an occupancy map that correspond to the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images that correspond to the particular object, wherein the occupancy map indicates portions of the one or more 2D image frames that are packed with patches and portions that are padded; and

a third entry comprising information defining a 3D volume space of the point cloud occupied by the particular object.

12. The device of claim 11 , wherein the program instructions, when executed on or across the one or more processors, cause the one or more processors to:

encode an additional supplementary message canceling or updating which ones of the attribute patch images and depth patch images of the one or more 2D image frames are included in the respective sub-sets that correspond to the particular object;

encode an additional supplementary message canceling or updating respective portions of the occupancy map that correspond to attribute and depth patch images of the one or more 2D image frames that correspond to the particular object; or

encode an additional supplementary message canceling or updating a 3D volume space of the point cloud occupied by the particular object.

13. The device of claim 12 , wherein the one or more supplementary messages further comprise:

an additional entry comprising an object ID for the particular object,

wherein the first entry identifying the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images of the one or more 2D image frames that correspond to the particular object is linked to the object ID,

wherein the second entry identifying the portions of the occupancy map that correspond to the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images that correspond to the particular object is linked to the object ID; and

wherein the third entry comprising the information defining the 3D volume space occupied by the particular object is linked to the object ID.

14. The device of claim 13 , wherein the program instructions, when executed one or across the one or more processors, cause the one or more processors to:

determine an updated characteristic for the particular object linked to the object ID; and

encode an update supplementary message that includes the object ID and information indicating the updated characteristic of the particular object without repeating the signaling of information previously signaled for the particular object linked to the object ID that is still applicable to the particular object.

15. The device of claim 13 , wherein the program instructions, when executed one or across the one or more processors, further cause the one or more processors to:

determine a location of the particular object in the point cloud at a subsequent moment in time; and

encode an additional supplementary message comprising information defining a 3D volume space of the point cloud occupied by the particular object at the subsequent moment in time,

wherein the additional supplementary message comprises the object ID for the particular object and does not repeat the first entry and the second entries previously encoded for the particular object in the supplementary message.

16. The device of claim 13 , wherein the program instructions, when executed on or across the one or more processors, cause the one or more processors to:

encode separate supplementary messages for:

the first or second entries comprising the information identifying the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images that correspond to the particular object or the information identifying the portions of the occupancy map that correspond to the sub-set of the packed attribute patch images and the sub-set of the packed depth patch images that correspond to the particular object; and

the third entry comprising the information defining the 3D volume space occupied by the particular object.

17. A non-transitory, computer-readable, medium storing program instructions, that when executed on or across one or more processors, cause the one or more processors to:

encode one or more supplementary messages for compressed visual volumetric content, the one or more supplementary messages comprising:

a first entry identifying a sub-set of packed attribute and depth patch images of one or more 2D image frames that correspond to a particular object of a plurality of objects in the compressed visual volumetric content;

a second entry identifying respective portions of an occupancy map that correspond to the sub-set of the packed attribute patch images and the depth patch images that correspond to the particular object, wherein the occupancy map indicates portions of the one or more 2D image frames that are packed with patches and portions that are padded; and

a third entry comprising information defining a 3D volume space of the visual volumetric content occupied by the particular object;

wherein the sub-set of the packed attribute patch images and depth patch images that correspond to the particular object is a sub-set that includes less than a total number of packed attribute patch images and depth patch images that correspond to the plurality of objects;

wherein one or more of the first entry, the second entry, or the third entry comprise respective patch identifiers for respective patches that correspond to portions of the one or more 2D image frames that comprise the sub-set of the set of packed attribute patch images that correspond to the particular object and the sub-set of the set of packed depth patch images that correspond to the particular object in the point cloud being encoded; and

provide the encoded one or more supplementary messages for use in decoding the compressed visual volumetric content.

18. The non-transitory, computer-readable, medium of claim 17 , wherein the one or more supplementary messages further comprise:

an additional entry comprising an object ID for the particular object,

wherein the first entry identifying the sub-set of the packed attribute patch images and depth patch images of the one or more 2D image frames that correspond to the particular object is linked to the object ID,

wherein the second entry identifying the portions of the occupancy map that correspond to the sub-set of the packed attribute patch images and the depth patch images corresponding to the particular object is linked to the object ID; and

wherein the third entry comprising the information defining the 3D volume space occupied by the particular object is linked to the object ID.

19. The non-transitory, computer-readable, medium of claim 18 , wherein the program instructions, when executed one or across the one or more processors, cause the one or more processors to:

determine an updated characteristic for the particular object linked to the object ID; and

encode an update supplementary message that includes the object ID and information indicating the updated characteristic of the particular object without repeating the signaling of information previously signaled for the particular object linked to the object ID that is still applicable to the particular object.

20. A method of compressing visual volumetric content, the method comprising:

generating, for respective sets of points of the visual volumetric content, respective attribute patch images and respective depth patch images comprising attribute information or geometry information for the respective sets of points projected onto respective patch planes;

packing a set of the generated attribute patch images and a set of the generated depth patch images in one or more 2D image frames;

video encoding the one or more 2D image frames; and

encoding one or more supplementary messages comprising one or more entries, wherein the one or more entries define a particular object of a plurality of objects included in the visual volumetric content and comprises respective patch identifiers for respective patches that correspond to portions of the one or more 2D image frames that comprise a sub-set of the set of packed attribute patch images that correspond to the particular object and a sub-set of the set of packed depth patch images that correspond to the particular object in the visual volumetric content being encoded, wherein the sub-set of packed attribute patch images that correspond to the particular object is a sub-set that includes less than a total number of packed attribute patch images that correspond to the plurality of objects, and wherein the sub-set of the set of packed depth patch images that correspond to the particular object is a sub-set that includes less than a total number of packed depth patch images that correspond to the plurality of objects.

Continuity (3)

References Cited (400)

Cited By (2)

US 12,530,812 US 12,694,473