IP Library Granted Patent US 9,424,663
Granted Patent B2
US 9,424,663 · App. 14/310,514 · Granted Aug 23, 2016

Apparatus and method for coding a three dimensional mesh

Inventors: Min Su Ahn (Seoul, KR); Jeong Hwan Ahn (Suwon-si, KR); Jae Kyun Ahn (Seoul, KR); Dae Youn Lee (Seoul, KR); Chang Su Kim (Seoul, KR)
Assignees: Samsung Electronics Co., Ltd.; Korea University Industrial & Academic Collaboration Foundation
G06T9/001G06T9/00G06T17/20G06T17/205
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,424,663
App. No.
14/310,514
Granted
Aug 23, 2016
Kind
B2
Abstract

Disclosed is a three-dimensional (3D) mesh compression apparatus and method. The 3D mesh compression apparatus may generate a base mesh through a mesh simplification, may separately compress the base mesh and vertices eliminated by the simplification, and may compress 3D mesh data based on the covariance matrix.

Claims (81)

1. A method of compressing a 3D mesh, the method comprising:

simplifying a mesh by eliminating a plurality of vertices; and

generating a covariance matrix using the plurality of vertices, the generating comprising evaluating at least one topological distance among the plurality of eliminated vertices.

2. The method of claim 1 further comprising:

partitioning the simplified mesh into a plurality of segments.

3. The method of claim 2 , wherein each of the plurality of segments contains less than the predefined number of vertices.

4. The method of claim 1 , wherein the covariance matrix is

M

=

[

1

ρ

N

1

j

ρ

N

1

m

ρ

N

ij

ρ

N

i

1

1

ρ

N

im

ρ

N

ji

ρ

N

m

1

ρ

N

mj

1

]

,

ρ denotes a correlation coefficient having a value equal or less than 1, and N ij denotes the at least one topological distance among the plurality of eliminated vertices.

5. The method of claim 1 further comprising:

calculating at least one eigenvector by performing eigen-vector decomposition of the covariance matrix.

6. The method of claim 1 further comprising:

calculating a coefficient of KL transform by using at least one eigenvector and at least one residual vector.

7. The method of claim 1 further comprising:

calculating a coefficient of KL transform by using y=U T x,

wherein x is a vector obtained from at least one residual vector, y is the coefficient of KL transform, and U has column vectors, which are the eigenvectors.

8. The method of claim 7 further comprising:

estimating the at least one residual vector.

9. The method of claim 7 further comprising:

quantizing the calculated coefficient of KL transform.

10. The method of claim 1 further comprising:

compressing the 3D mesh by using the covariance matrix.

11. A method of compressing a 3D mesh, the method comprising:

generating a simplified mesh by simplifying an inputted 3D mesh, thereby eliminating at least one vertex;

generating a covariance matrix using the at least one vertex; and

coding the simplified mesh thereby generating a bitstream comprising information regarding the simplified mesh, the information comprising connectivity information of the simplified mesh and geometric information of the simplified mesh.

Priority Claims (2)
KR 10-2009-0098952 · Oct 16, 2009 · national
KR 10-2010-0034488 · Apr 14, 2010 · national
Continuity (2)
Continuation 12905607 · Oct 15, 2010
Related Publication 20140300598A1 · Oct 9, 2014