IP Library Granted Patent US 10,657,675
Granted Patent B2
US 10,657,675 · App. 15/913,824 · Granted May 19, 2020

Techniques for improved progressive mesh compression

Inventors: Michael Hemmer (San Francisco, CA); Pierre Alliez (Sophia-Antipolis, FR); Cedric Portaneri (Sophia-Antipolis, FR)
Assignee: GOOGLE LLC
G06T9/001G06T2210/36
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 10,657,675
App. No.
15/913,824
Granted
May 19, 2020
Kind
B2
Abstract

An encoder includes a processor and a memory. The encoder generates a first plurality of levels of detail (LODs) and associated first type of vertex split records, each of the first type of vertex split records associated with an LOD of the first plurality of LODs is generated using a first type of collapse operator. The encoder initiates a switch from using the first type of collapse operator to a second type of collapse operator in response to a switching condition being satisfied. The encode further a second plurality of LODs and associated second type of vertex split records, each of the second type of vertex split records associated with a LOD of the second plurality of LODs is generated using the second type of collapse operator.

Claims (47)

1. A computer-implemented method of progressive mesh compression, comprising:

generating, at an encoder, a first plurality of levels of detail (LODs) and associated first type of vertex split records, the first type of vertex split records associated with an LOD of the first plurality of LODs are generated using a first type of collapse operator;

switching from using the first type of collapse operator to a second type of collapse operator in response to a switching condition being satisfied, the switching condition being based at least on identifying a branching point, to balance distortion and bits per vertex, from a plurality of branching points; and

generating, at the encoder, after the switching, a second plurality of LODs and associated second type of vertex split records, the second type of vertex split records associated with a LOD of the second plurality of LODs are generated using the second type of collapse operator.

2. The method of claim 1 , further comprising:

transmitting, from the encoder, in response to a message received from a decoder, a LOD of the first plurality of LODs and associated first type of vertex split records or a LOD of the second plurality of LODs and associated second type of vertex split records, wherein the LOD is a lowest LOD being requested from the decoder, and the vertex split records include vertex split records associated with the lowest LOD and vertex split records associated with LODs up to a highest LOD.

3. The method of claim 1 , wherein generating a LOD of the first plurality of LODs and associated first type of vertex split records includes:

collapsing, using the first type of collapse operator, edges of a LOD immediately preceding the LOD being generated; and

generating the associated vertex split records based on the collapsing.

4. The method of claim 1 , wherein generating a LOD of the second plurality of LODs and associated second type of vertex split records includes:

collapsing, using the second type of collapse operator, edges of a LOD immediately preceding the LOD being generated; and

generating the associated vertex split records based on the collapsing.

5. The method of claim 1 , wherein the first type of collapse operator is a full-edge collapse operator and the second type of collapse operator is a half-edge collapse operator.

6. The method of claim 1 , wherein the switching condition is based at least on one of a user-defined parameter or an error threshold.

7. The method of claim 1 , wherein the switching condition is based at least on identifying a branching point upon generating of the first LOD, the switching being triggered at the branching point.

8. The method of claim 1 , wherein the first type of collapse operator is a half-edge collapse operator and the second type of collapse operator is a full-edge collapse operator.

9. An encoder, comprising:

a processor; and

a memory, the memory including instructions configured to cause the processor to:

generate a first plurality of levels of detail (LODs) and associated first type of vertex split records, the first type of vertex split records associated with an LOD of the first plurality of LODs are generated using a first type of collapse operator;

switch from using the first type of collapse operator to a second type of collapse operator in response to a switching condition being satisfied, the switching condition being based at least on identifying a branching point, to balance distortion and bits per vertex, from a plurality of branching points; and

generate, after the switching, a second plurality of LODs and associated second type of vertex split records, the second type of vertex split records associated with a LOD of the second plurality of LODs are generated using the second type of collapse operator.

10. The encoder of claim 9 , further comprising:

transmitting, from the encoder, in response to a message received from a decoder, a LOD of the first plurality of LODs and associated first type of vertex split records or a LOD of the second plurality of LODs and associated second type of vertex split records, wherein the LOD is a lowest LOD being requested from the decoder, and the vertex split records include vertex split records associated with the lowest LOD and vertex split records associated with LODs up to a highest LOD.

11. The encoder of claim 9 , wherein generating a LOD of the first plurality of LODs and associated first type of vertex split records includes:

collapsing, using the first type of collapse operator, edges of a LOD immediately preceding the LOD being generated; and

generating the associated vertex split records based on the collapsing.

12. The encoder of claim 9 , wherein generating a LOD of the second plurality of LODs and associated second type of vertex split records includes:

collapsing, using the second type of collapse operator, edges of a LOD immediately preceding the LOD being generated; and

generating the associated vertex split records based on the collapsing.

13. The encoder of claim 9 , wherein the first type of collapse operator is a full-edge collapse operator and the second type of collapse operator is a half-edge collapse operator.

14. The encoder of claim 9 , wherein the switching condition is based at least on one of a user-defined parameter or an error threshold.

15. The encoder of claim 9 , wherein the switching condition is based at least on identifying a branching point upon generating of the first LOD, the switching being triggered at the branching point.

16. A non-transitory computer-readable storage medium having stored thereon computer executable program code which, when executed on a computer system, causes the computer system to perform a method, comprising:

generating, at an encoder, a first plurality of levels of detail (LODs) and associated first type of vertex split records, the first type of vertex split records associated with an LOD of the first plurality of LODs are generated using a first type of collapse operator;

switching from using the first type of collapse operator to a second type of collapse operator in response to a switching condition being satisfied, the switching condition being based at least on identifying a branching point, to balance distortion and bits per vertex, from a plurality of branching points; and

generating, at the encoder, after the switching, a second plurality of LODs and associated second type of vertex split records, the second type of vertex split records associated with a LOD of the second plurality of LODs are generated using the second type of collapse operator.

17. The computer-readable storage medium of claim 16 , the method further comprising:

transmitting, from the encoder, in response to a message received from a decoder, a LOD of the first plurality of LODs and associated first type of vertex split records or a LOD of the second plurality of LODs and associated second type of vertex split records, wherein the LOD is a lowest LOD being requested from the decoder, and the vertex split records include vertex split records associated with the lowest LOD and vertex split records associated with LODs up to a highest LOD.

18. The computer-readable storage medium of claim 16 , wherein generating a LOD of the first plurality of LODs and associated first type of vertex split records includes:

collapsing, using the first type of collapse operator, edges of a LOD immediately preceding the LOD being generated; and

generating the associated vertex split records based on the collapsing.

19. The computer-readable storage medium of claim 16 , wherein generating a LOD of the second plurality of LODs and associated second type of vertex split records includes:

collapsing, using the second type of collapse operator, edges of a LOD immediately preceding the LOD being generated; and

generating the associated vertex split records based on the collapsing.

20. The computer-readable storage medium of claim 16 , wherein the first type of collapse operator is a full-edge collapse operator and the second type of collapse operator is a half-edge collapse operator.

21. The computer-readable storage medium of claim 16 , wherein the switching condition is based at least on one of a user-defined parameter or an error threshold.

Assignments (2)
CONFIRMATORY ASSIGNMENT Recorded Mar 10, 2023
From: INRIA CENTRE AT UNIVERSITÉ CÔTE D'AZUR
To: GOOGLE LLC
Reel/Frame 063026/0073 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Mar 23, 2018
From: HEMMER, MICHAEL; ALLIEZ, PIERRE; PORTANERI, CEDRIC
To: GOOGLE LLC
Reel/Frame 045322/0344 →
Continuity (1)
Related Publication 20190279422A1 · Sep 12, 2019