IP Library Granted Patent US 10,867,395
Granted Patent B2
US 10,867,395 · App. 16/147,107 · Granted Dec 15, 2020

Systems and methods for real-time rigid body motion prediction

Inventors: Kim Wei Ng (Yuen Long, HK); Ka Lok Ng (Kowloon, HK)
Assignee: Glo Big Boss Ltd.
G06T7/246G06K9/00342G06T7/292G06T7/73G06T13/20G06T2207/10048G06T2207/30204
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 10,867,395
App. No.
16/147,107
Granted
Dec 15, 2020
Kind
B2
Abstract

Methods and systems for real-time rigid body motion prediction are provided. The method includes determining an occurrence of an untracked rigid body in a current frame, the untracked rigid body being tracked in a previous frame. The method also includes predicting a position in the current frame of the untracked rigid body using previously tracked data in the previous frame. The method also includes identifying unlabeled markers in the current frame that are proximate to the predicted position of the untracked rigid body. The method also includes reconstructing the untracked rigid body in the current frame based on the identifying of the unlabeled markers.

Claims (55)

1. A computer-implemented method for rigid body motion prediction comprising:

tracking movements of at least one rigid body;

determining an occurrence of an untracked rigid body in a current frame, the untracked rigid body comprising a rigid body that was tracked in a previous frame;

predicting a position in the current frame of the untracked rigid body using previously tracked data of the untracked rigid body in the previous frame;

identifying unlabeled markers in the current frame that are proximate to the predicted position of the untracked rigid body in the current frame; and

reconstructing the untracked rigid body in the current frame based on the identifying.

2. The computer-implemented method of claim 1 , further comprising:

predicting a current orientation in the current frame of the untracked rigid body using the previously tracked data in the previous frame.

3. The computer-implemented method of claim 1 , wherein the predicting further comprises:

interpolating angular velocity, angular acceleration, linear velocity, and/or linear acceleration of the untracked rigid body using the previously tracked data.

4. The computer-implemented method of claim 1 , further comprising:

labeling the unlabeled markers corresponding to the untracked rigid body.

5. The computer-implemented method of claim 1 , further comprising:

comparing currently tracked rigid bodies in the current frame with previously tracked rigid bodies in the previous frame.

6. The computer-implemented method of claim 5 , further comprising:

defining a maximum radius about each previously tracked rigid body.

7. The computer-implemented method of claim 6 , further comprising:

filtering out the untracked rigid body when a distance between the untracked rigid body and the previously tracked rigid body exceeds the maximum radius.

8. The computer-implemented method of claim 1 , wherein each previously tracked rigid body comprises labeled markers, further comprising:

defining a radius around each labeled marker.

9. The computer-implemented method of claim 8 , further comprising:

filtering out the unlabeled markers when a distance between the unlabeled markers and the labeled markers exceeds the radius.

10. The computer-implemented method of claim 1 , wherein the reconstructing further comprises:

defining at least one vector based on the unlabeled markers; and

defining a pivot point using the previously tracked data to reconstruct a rigid body pose.

11. The computer-implemented method of claim 10 , further comprising:

determining a midpoint between the unlabeled markers.

12. A system for rigid body motion prediction, comprising:

a memory storing sequences of instructions; and

a processor configured to execute the sequences of instructions which, when executed, causes the processor to perform:

tracking movements of at least one rigid body;

determining an occurrence of an untracked rigid body in a current frame, the untracked rigid body comprising a rigid body that was tracked in a previous frame;

predicting a position in the current frame of the untracked rigid body using previously tracked data of the untracked rigid body in the previous frame;

identifying unlabeled markers in the current frame that are proximate to the predicted position of the untracked rigid body in the current frame; and

reconstructing the untracked rigid body in the current frame based on the identifying.

13. The system of claim 12 , further comprising stored sequences of instructions, which when executed by the processor, cause the processor to perform:

predicting a current orientation in the current frame of the untracked rigid body using previously tracked data in the previous frame.

14. The system of claim 12 , further comprising stored sequences of instructions, which when executed by the processor, cause the processor to perform:

interpolating angular velocity, angular acceleration, linear velocity, and/or linear acceleration of the untracked rigid body using previously tracked data in the previous frame.

15. The system of claim 12 , further comprising stored sequences of instructions, which when executed by the processor, cause the processor to perform:

labeling the unlabeled markers corresponding to the untracked rigid body.

16. The system of claim 12 , further comprising stored sequences of instructions, which when executed by the processor, cause the processor to perform:

comparing currently tracked rigid bodies in the current frame with previously tracked rigid bodies in the previous frame.

17. The system of claim 16 , further comprising stored sequences of instructions, which when executed by the processor, cause the processor to perform:

defining a maximum radius about each previously tracked rigid body.

18. The system of claim 17 , further comprising stored sequences of instructions, which when executed by the processor, cause the processor to perform:

filtering out the untracked rigid body when a distance between the untracked rigid body and the previously tracked rigid body exceeds the maximum radius.

19. A non-transitory computer readable storage medium is provided including instructions that, when executed by a processor, case the processor to perform a method for rigid body motion prediction, the method comprising:

tracking movements of at least one rigid body;

determining an occurrence of an untracked rigid body in a current frame, the untracked rigid body comprising a rigid body that was tracked in a previous frame;

predicting a position in the current frame of the untracked rigid body using previously tracked data of the untracked rigid body in the previous frame;

identifying unlabeled markers in the current frame that are proximate to the predicted position of the untracked rigid body in the current frame; and

reconstructing the untracked rigid body in the current frame based on the identifying.

20. The non-transitory computer readable storage medium of claim 19 , wherein the method further comprises:

predicting a current orientation in the current frame of the untracked rigid body using the previously tracked data in the previous frame.

Assignments (7)
SECURITY INTEREST Recorded Dec 18, 2025
From: SANDBOX VR, INC.; GLOSTATION USA, INC.; GLOSTATION CORE USA, INC.; GLOSTATION FRANCHISING USA, INC.; GLO BIG BOSS LIMITED
To: TRIPLEPOINT CAPITAL LLC
Reel/Frame 073264/0636 →
RELEASE OF SECURITY INTEREST Recorded Oct 19, 2022
From: SILICON VALLEY BANK
To: GLO BIG BOSS LTD.
Reel/Frame 061472/0273 →
SECURITY INTEREST Recorded Aug 27, 2020
From: SANDBOX VR, INC.; GLOSTATION USA, INC.; GLOSTATION CORE USA, INC.; GLOSTATION FRANCHISING USA, INC.; SANDBOX VR TOPANGA, LLC; SANDBOX VR MISSION VALLEY, LLC; SANDBOX VR SAN MATEO, LLC; SANDBOX VR CERRITOS, LLC; SANDBOX VR RIDGE HILL, LLC; SANDBOX VR AUSTIN, LLC; SANDBOX VR COLONY, LLC; SANDBOX VR OAKBROOK, LLC; SANDBOX VR POP-UP, LLC; GLO BIG BOSS LIMITED
To: TRIPLEPOINT CAPITAL LLC
Reel/Frame 053622/0554 →
INTELLECTUAL PROPERTY SECURITY AGREEMENT Recorded Aug 20, 2020
From: GLO BIG BOSS LTD.
To: SILICON VALLEY BANK
Reel/Frame 053568/0545 →
CORRECTIVE ASSIGNMENT TO CORRECT THE APPLICATION NUMBER 14147107 PREVIOUSLY RECORDED AT REEL: 047502 FRAME: 0876. ASSIGNOR(S) HEREBY CONFIRMS THE ASSIGNMENT . Recorded May 23, 2019
From: NG, KIM WEI; NG, KA LOK
To: GLO BIG BOSS LIMITED
Reel/Frame 049283/0690 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 13, 2018
From: NG, KIM WEI; NG, KA LOK
To: GLO BIG BOSS LIMITED
Reel/Frame 047512/0523 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 12, 2018
From: NG, KIM WEI; NG, KA LOK
To: GLO BIG BOSS LIMITED
Reel/Frame 047502/0876 →
Continuity (1)
Related Publication 20200105000A1 · Apr 2, 2020