Patch splitting for improving video-based point cloud compression performance

1. An encoding device for point cloud encoding, the encoding device comprising:

a processor configured to:

generate, for a three-dimensional (3D) point cloud, a first frame and a second frame that include patches representing a cluster of points of the 3D point cloud;

identify a first patch to segment in the patches of the first frame and the second frame;

identify an attribute for points in the first patch;

determine, in response to identifying the first patch, transition points in the first patch based on the attribute changing between adjacent points;

identify an attribute boundary based on a string of the transition points extending entirely across the first patch;

determine a path representing the attribute boundary between segmented regions within the patch;

segment the first patch along the path into a second patch and a third patch for the first frame and the second frame; and

encode the second patch and the third patch in the first frame and the second frame to generate a compressed bitstream; and

a communication interface operably coupled to the processor, the communication interface configured to transmit the compressed bitstream.

2. The encoding device of claim 1 , wherein:

the processor is further configured to identify the first patch to segment based on a dimension of the first patch exceeding a dimension threshold, and

the determination of the path is in response to the identification that the dimension of the first patch exceeds the dimension threshold.

3. The encoding device of claim 2 , wherein:

more than one attribute boundary is identified based on the transition points,

to determine the path, the processor is further configured to determine the path based on an attribute boundary with a longest boundary in a direction that the first patch exceeded the dimension threshold.

4. The encoding device of claim 2 , wherein the dimension threshold is a max patch size.

5. The encoding device of claim 2 , wherein the dimension threshold is a function of a point cloud bit-depth.

6. The encoding device of claim 2 , wherein, to determine the path, the processor is further configured to determine that at least one of the second patch and the third patch does not exceed the dimension threshold, which was exceeded by the first patch.

7. The encoding device of claim 2 , wherein:

the attribute is a texture; and

the attribute boundary is a texture boundary.

8. The encoding device of claim 2 , wherein:

the attribute is a depth, and

the attribute boundary is a depth boundary.

9. The encoding device of claim 2 , wherein:

the attribute is a material identifier (ID); and

the attribute boundary is a material boundary.

10. The encoding device of claim 1 , wherein the path continuously extends from one side of the first patch to an opposite side of the first patch in a direction in which the first patch exceeded a dimension threshold.

11. A method for point cloud encoding, the method comprising:

generating, using a processor of an encoder, a first frame and a second frame that include patches representing a cluster of points of three-dimensional (3D) point cloud;

identifying a first patch to segment in the patches of the first frame and the second frame;

identifying an attribute for points in the first patch;

determining, in response to identifying the first patch, transition points in the first patch based on the attribute changing between adjacent points;

identifying an attribute boundary based on a string of the transition points extending entirely across the first patch;

determining a path representing the attribute boundary between segmented regions within the patch;

segmenting the first patch along the path into a second patch and a third patch for the first frame and the second frame;

encoding the second patch and the third patch in the first frame and the second frame to generate a compressed bitstream; and

transmitting, using a communication interface operably coupled to the processor, the compressed bitstream.

12. The method of claim 11 , further comprising identifying the first patch to segment based on a dimension of the first patch exceeding a dimension threshold,

wherein the determination of the path is in response to the identification that the dimension of the first patch exceeds the dimension threshold.

13. The method of claim 12 , wherein:

more than one attribute boundary is identified based on the transition points,

to determine the path, the method further comprises determining the path based on an attribute boundary with a longest boundary in a direction that the first patch exceeded the dimension threshold.

14. The method of claim 12 , wherein the dimension threshold is a max patch size.

15. The method of claim 12 , wherein the dimension threshold is a function of a point cloud bit-depth.

16. The method of claim 12 , wherein determining the path includes determining that at least one of the second patch and the third patch does not exceed the dimension threshold, which was exceeded by the first patch.

17. The method of claim 12 , wherein:

the attribute is a texture; and

the attribute boundary is a texture boundary.

18. The method of claim 12 , wherein:

the attribute is a depth, and

the attribute boundary is a depth boundary.

19. The method of claim 12 , wherein:

the attribute is a material identifier (ID); and

the attribute boundary is a material boundary.

20. The method of claim 11 , wherein the path continuously extends from one side of the first patch to an opposite side of the first patch in a direction in which the first patch exceeded a dimension threshold.