Granted Patent B2

US 12,299,941 · App. 17/903,505 · Granted May 13, 2025

Inventors: Chao Huang (Palo Alto, CA); Xiang Zhang (Sunnyvale, CA); Jun Tian (Belle Mead, NJ); Xiaozhong Xu (State College, PA); Shan Liu (San Jose, CA)

Assignee: TENCENT AMERICA LLC

G06T9/001G06T17/20

Abstract

Claims (65)

1. A method for mesh decompression, comprising:

decoding an array of attributes from a bitstream carrying a three dimensional (3D) mesh frame that includes a plurality of patches, the array of attributes corresponding to vertices of the 3D mesh frame, the vertices of the 3D mesh frame being ordered into subsets respectively belonging to the plurality of patches;

selecting a first portion of the array of attributes corresponding to a first subset of the vertices of the 3D mesh frame based on a first value decoded from the bitstream that is indicative of a total number of vertices in the first subset of the vertices of the 3D mesh frame, the first subset belonging to a first patch;

selecting attributes of boundary vertices in the first subset from the first portion of the array of attributes based on a second value, the second value being decoded from the bitstream and indicative of a total number of the boundary vertices in the first subset, the attributes of the boundary vertices being ordered before attributes of other vertices in the first subset;

determining first connectivity information of the first subset of the vertices; and

reconstructing the first patch of the 3D mesh frame based on the first portion of the array of attributes and the first connectivity information.

2. The method of claim 1 , wherein the decoding the array of attributes further comprises at least one of:

decoding a 2 dimensional (2D) image corresponding to an array of 3 dimensional coordinates of the vertices;

decoding a 2D image corresponding to an array of uv coordinates of the vertices; and/or

decoding a first 2D image and a second 2D image respectively corresponding to a first array of u coordinates of the vertices and a second array of v coordinates of the vertices.

3. The method of claim 1 , wherein the attributes of the boundary vertices comprise coordinates of the boundary vertices, and the method further comprises:

decoding boundary coordinate residuals for the boundary vertices; and

combining the coordinates of the boundary vertices with the boundary coordinate residuals of the boundary vertices to obtain recovered coordinates of the boundary vertices.

4. The method of claim 3 , wherein the coordinates of the boundary vertices comprises at least one of 3 dimensional (3D) coordinates of the boundary vertices and/or uv coordinates of the boundary vertices, and the boundary coordinate residuals for the boundary vertices comprises at least one of boundary 3D coordinates residuals and/or boundary uv coordinates residuals.

5. The method of claim 3 , further comprising:

detecting that a first vertex and a second vertex have identical coordinates; and

removing one of the first vertex and the second vertex from the reconstructing.

6. The method of claim 3 , further comprising:

detecting that a non-boundary vertex of the first patch is outside a boundary formed by the boundary vertices of the first patch; and

reconstructing the first patch of the 3D mesh frame without the non-boundary vertex.

7. The method of claim 1 , wherein the determining the first connectivity information of the first subset of the vertices further comprises:

inferring the first connectivity information of the first subset of the vertices.

8. The method of claim 1 , wherein the determining the first connectivity information of the first subset of the vertices further comprises:

decoding, from the bitstream, the first connectivity information of the first subset of the vertices.

9. A method for mesh encoding, comprising:

encoding an array of attributes of a three-dimensional (3D) mesh frame that includes a plurality of patches, the array of attributes corresponding to vertices of the 3D mesh frame, the vertices of the 3D mesh frame being ordered into subsets respectively belonging to the plurality of patches;

selecting a first portion of the array of attributes corresponding to a first subset of the vertices of the 3D mesh frame based on a first value that is indicative of a total number of vertices in the first subset, the first subset belonging to a first patch;

selecting attributes of boundary vertices in the first subset from the first portion of the array of attributes based on a second value indicative of a total number of the boundary vertices in the first subset, the attributes of the boundary vertices being ordered before attributes of other vertices in the first subset;

determining first connectivity information of the first subset of the vertices; and

encoding the first patch of the 3D mesh frame based on the first portion of the array of attributes and the first connectivity information.

10. The method of claim 9 , wherein the encoding the array of attributes further comprises at least one of:

encoding a 2-dimensional (2D) image corresponding to an array of 3 dimensional coordinates of the vertices;

encoding a 2D image corresponding to an array of uv coordinates of the vertices; and/or

encoding a first 2D image and a second 2D image respectively corresponding to a first array of u coordinates of the vertices and a second array of v coordinates of the vertices.

11. The method of claim 9 , further comprising:

encoding the first value that is indicative of the total number of vertices in the first subset of the vertices of the 3D mesh frame.

12. The method of claim 9 , further comprising: encoding the second value indicative of the total number of the boundary vertices in the first subset.

13. The method of claim 9 , wherein the attributes of the boundary vertices include coordinates of the boundary vertices, and the method further comprises:

encoding boundary coordinate residuals for the boundary vertices.

14. The method of claim 13 , wherein the coordinates of the boundary vertices comprises at least one of 3 dimensional (3D) coordinates of the boundary vertices and/or uv coordinates of the boundary vertices, and the boundary coordinate residuals for the boundary vertices comprises at least one of boundary 3D coordinates residuals and/or boundary uv coordinates residuals.

15. The method of claim 13 , further comprising:

detecting that a first vertex and a second vertex have identical coordinates; and

removing one of the first vertex and the second vertex from the first patch of the 3D mesh frame to be encoded.

16. The method of claim 13 , further comprising:

detecting that a non-boundary vertex of the first patch is outside a boundary formed by the boundary vertices of the first patch; and

encoding the first patch of the 3D mesh frame without the non-boundary vertex.

17. The method of claim 9 , wherein the determining the first connectivity information of the first subset of the vertices further comprises:

inferring the first connectivity information of the first subset of the vertices.

18. A method of processing mesh data, the method comprising:

processing a bitstream that includes the mesh data according to a format rule, wherein:

the bitstream includes an array of attributes carrying a three-dimensional (3D) mesh frame that includes a plurality of patches, the array of attributes corresponding to 3D mesh frame, the vertices of the 3D mesh frame being ordered into subsets respectively belonging to the plurality of patches; and

the format rule specifies that:

a first portion of the array of attributes corresponding to a first subset of the vertices of the 3D mesh frame is selected based on a first value decoded from the bitstream that is indicative of a total number of vertices in the first subset of the vertices of the 3D mesh frame, the first subset belonging to a first patch,

attributes of boundary vertices in the first subset are selected from the first portion of the array of attributes based on a second value, the second value being decoded from the bitstream and indicative of a total number of the boundary vertices in the first subset, the attributes of the boundary vertices being ordered before attributes of other vertices in the first subset,

first connectivity information of the first subset of the vertices is determined, and

the first patch of the 3D mesh frame is processed based on the first portion of the array of attributes and the first connectivity information.

19. The method of claim 18 , wherein the format rule specifies that:

a 2 dimensional ( 2 D) image corresponding to an array of 3 dimensional coordinates of the vertices is decoded;

a 2D image corresponding to an array of uv coordinates of the vertices is decoded; and/or

a first 2D image corresponding to a first array of u coordinates of the vertices and a second 2D image corresponding to a second array of v coordinates of the vertices are decoded respectively.

20. The method of claim 18 , wherein:

the attributes of the boundary vertices include coordinates of the boundary vertices; and

the format rule specifies that:

boundary coordinate residuals are decoded for the boundary vertices; and

the coordinates of the boundary vertices are combined with the boundary coordinate residuals of the boundary vertices to obtain recovered coordinates of the boundary vertices.

Assignments (1)

Continuity (3)

References Cited (25)