Granted Patent B2

US 12,626,413 · App. 18/486,784 · Granted May 12, 2026

Inventors: Rajan Laxman Joshi (San Diego, CA); Madhukar Budagavi (Plano, TX)

Assignee: Samsung Electronics Co., Ltd.

G06T9/001H04N19/119H04N19/172H04N19/184H04N19/597

Abstract

Claims (46)

1 . An apparatus comprising:

a communication interface; and

a processor operably coupled to the communication interface, the processor configured to:

subdivide an intermediate mesh-frame, created from a reconstructed first displacement field and a first subdivided mesh-frame, to create a second subdivided mesh-frame;

extract a set of vertex positions from the second subdivided mesh-frame and generate a second displacement field based on the second subdivided mesh-frame and an original mesh-frame, wherein the extracted set of vertex positions of the second subdivided mesh-frame comprise only new vertex positions generated by the subdivision of the intermediate mesh-frame; and

encode the second displacement field into a bitstream.

2 . The apparatus of claim 1 , wherein, to create the intermediate mesh-frame, the processor is further configured to:

decode an encoded first displacement field to obtain the reconstructed first displacement field; and

combine the reconstructed first displacement field with the first subdivided mesh-frame to create the intermediate mesh-frame.

3 . The apparatus of claim 2 , wherein the processor is further configured to:

subdivide a reconstructed base mesh-frame to create the first subdivided mesh-frame, wherein the reconstructed base mesh-frame is associated with the original mesh-frame;

generate a first displacement field based on the first subdivided mesh-frame and the original mesh-frame; and

encode the first displacement field to obtain the encoded first displacement field and write the encoded first displacement field into the bitstream.

4 . The apparatus of claim 3 , wherein the reconstructed base mesh-frame is created by decoding an encoding of an original base mesh-frame.

5 . The apparatus of claim 1 , wherein, to generate the second displacement field, the processor is further configured to:

determine a difference between the extracted set of vertex positions of the second subdivided mesh-frame and a set of nearest vertex positions of the original mesh-frame.

6 . A method comprising:

subdividing an intermediate mesh-frame, created from a reconstructed first displacement field and a first subdivided mesh-frame, to create a second subdivided mesh-frame;

extracting a set of vertex positions from the second subdivided mesh-frame and generating a second displacement field based on the second subdivided mesh-frame and an original mesh-frame, wherein the extracted set of vertex positions of the second subdivided mesh-frame comprise only new vertex positions generated by the subdivision of the intermediate mesh-frame; and

encoding the second displacement field into a bitstream.

7 . The method of claim 6 , further comprising:

decoding an encoded first displacement field to obtain the reconstructed first displacement field; and

combining the reconstructed first displacement field with the first subdivided mesh-frame to create the intermediate mesh-frame.

8 . The method of claim 7 , further comprising:

subdividing a reconstructed base mesh-frame to create the first subdivided mesh-frame, wherein the reconstructed base mesh-frame is associated with the original mesh-frame;

generating a first displacement field based on the first subdivided mesh-frame and the original mesh-frame; and

encoding the first displacement field to obtain the encoded first displacement field and writing the encoded first displacement field into the bitstream.

9 . The method of claim 8 , wherein the reconstructed base mesh-frame is created by decoding an encoding of an original base mesh-frame.

10 . The method of claim 6 , wherein, to generate the second displacement field, the method further comprises:

determining a difference between the extracted set of vertex positions of the second subdivided mesh-frame and a set of nearest vertex positions of the original mesh-frame.

11 . An apparatus comprising:

a communication interface; and

a processor operably coupled to the communication interface, the processor configured to:

receive a compressed bitstream of a plurality of displacement fields;

decode the compressed bitstream and reconstruct the plurality of displacement fields; and

combine each one of the reconstructed plurality of displacement fields with one of a plurality of subdivided mesh-frames to reconstruct a mesh-frame, wherein the processor is further configured to create a second subdivided mesh-frame using a first subdivided mesh-frame of the plurality of subdivided mesh-frames and extract a set of vertex positions from the second subdivided mesh-frame, wherein the extracted set of vertex positions of the second subdivided mesh-frame comprise only new vertex positions generated by a subdivision of a first intermediate mesh-frame.

12 . The apparatus of claim 11 , wherein:

each one of the plurality of displacement fields is associated with one of a plurality of subdivision levels; and

each one of the plurality of subdivided mesh-frames is associated with one of the plurality of subdivision levels.

13 . The apparatus of claim 11 , wherein, to combine each one of the reconstructed plurality of displacement fields with one of the plurality of subdivided mesh-frames to reconstruct the mesh-frame, the processor is further configured to:

combine a reconstructed first displacement field of the reconstructed plurality of displacement fields with the first subdivided mesh-frame of the plurality of subdivided mesh-frames to create the first intermediate mesh-frame, wherein the first subdivided mesh-frame is created from a reconstructed base mesh-frame decoded from the compressed bitstream;

subdivide the first intermediate mesh-frame to create the second subdivided mesh-frame of the plurality of subdivided mesh-frames;

combine a reconstructed second displacement field of the reconstructed plurality of displacement fields with the second subdivided mesh-frame to create a second intermediate mesh-frame; and

combine the first intermediate mesh-frame and the second intermediate mesh-frame.

14 . The apparatus of claim 13 , wherein, to combine the reconstructed second displacement field of the reconstructed plurality of displacement fields with the second subdivided mesh-frame to create the second intermediate mesh-frame, the processor is further configured to:

add the reconstructed second displacement field to the extracted set of vertex positions of the second subdivided mesh-frame.

Assignments (1)

Continuity (2)

References Cited (9)