IP Library Granted Patent US 11,688,164
Granted Patent B2
US 11,688,164 · App. 17/323,461 · Granted Jun 27, 2023

Systems and methods for aligning map data

Inventors: Patrick Dubroy (Munich, DE); Wolfgang Hess (Munich, DE); Clemens Marschner (Munich, DE); Emil Praun (Union City, CA); Holger Rapp (Munich, DE); Thomas Schiwietz (Munich, DE); Vinay Shet (Fremont, CA)
Assignee: Woven Planet North America, Inc.
G06V20/20G06F16/29G06T7/70G06T17/05G06V20/56G06T2207/30256G06T2207/30261
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 11,688,164
App. No.
17/323,461
Granted
Jun 27, 2023
Kind
B2
Abstract

Systems, methods, and non-transitory computer-readable media can receive a geometric map and a semantic map associated with a geographic area, the semantic map comprising semantic data associated with vehicle navigation. A first semantic position estimate associated with a first piece of semantic data contained in the semantic map is generated based on semantic data location information associated with the first piece of semantic data. A final position for the first semantic position estimate is received. One or more three-dimensional semantic labels are applied to the geometric map based on the final position of the first semantic position estimate. A warped semantic map is generated. Generating the warped semantic map comprises warping the semantic map based on the one or more three-dimensional semantic labels.

Claims (40)

1. A computer-implemented method comprising:

receiving, by a computing system, a geometric map and a semantic map associated with a geographic area, the semantic map comprising semantic data associated with vehicle navigation;

generating, by the computing system, a semantic position estimate associated with an object represented in the semantic map based on semantic data location information associated with the object;

aligning the semantic position estimate with a physical feature that corresponds with the object to provide, by the computing system, a final position for the semantic position estimate that is a two-dimensional position;

applying, by the computing system, one or more three-dimensional semantic labels to the geometric map based on the final position of the semantic position estimate;

generating, by the computing system, a warped semantic map by warping the semantic map based on the one or more three-dimensional semantic labels; and

estimating a position of the object within an image associated with the geometric map by applying the semantic position estimate to the image, wherein aligning the semantic position estimate includes receiving the final position for the semantic position estimate, including the two-dimensional position within the image, and wherein applying the one or more three-dimensional semantic labels comprises translating the final position of the semantic position estimate within the image into a three-dimensional position within the geometric map.

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

receiving a set of images, including the image, associated with the geometric map.

3. The computer-implemented method of claim 2 , wherein receiving the final position for the semantic position estimate comprises:

receiving a user input adjusting a position of the semantic position estimate within the image.

4. The computer-implemented method of claim 2 , wherein receiving the final position for the semantic position estimate comprises:

adjusting a position of the semantic position estimate within the image using a machine learning model.

5. The computer-implemented method of claim 2 , wherein the final position of the semantic position estimate within the image is translated into a three-dimensional position within the geometric map based on image location information associated with the image and camera information associated with the image, and wherein the camera information comprises camera position, orientation, and direction information for a camera when the image was captured.

6. The computer-implemented method of claim 1 , wherein warping the semantic map based on the one or more three-dimensional semantic labels includes executing an optimization problem that warps the semantic map based on a set of constraints.

7. The computer-implemented method of claim 1 , further comprising layering the warped semantic map onto the geometric map.

8. A system comprising:

at least one processor; and

a memory storing instructions that, when executed by the at least one processor, cause the system to perform:

receiving a geometric map and a semantic map associated with a geographic area, the semantic map comprising semantic data associated with vehicle navigation;

generating a semantic position estimate associated with an object represented in the semantic map based on semantic data location information associated with the object;

aligning the semantic position estimate with a physical feature that corresponds with the object to provide a final position for the semantic position estimate that is a two-dimensional position;

applying one or more three-dimensional semantic labels to the geometric map based on the final position of the semantic position estimate;

generating a warped semantic map by warping the semantic map based on the one or more three-dimensional semantic labels; and

estimating a position of the object within an image associated with the geometric map by applying the semantic position estimate to the image, wherein aligning the semantic position estimate includes receiving the final position for the semantic position estimate, including the two-dimensional position within the image, and wherein applying the one or more three-dimensional semantic labels comprises translating the final position of the semantic position estimate within the image into a three-dimensional position within the geometric map.

9. The system of claim 8 , wherein the instructions further include instructions for:

receiving a set of images, including the image, associated with the geometric map.

10. The system of claim 9 , wherein receiving the final position for the semantic position estimate comprises:

receiving a user input adjusting a position of the semantic position estimate within the image.

11. A non-transitory computer-readable storage medium including instructions that, when executed by at least one processor of a computing system, cause the computing system to perform a method comprising:

receiving a geometric map and a semantic map associated with a geographic area, the semantic map comprising semantic data associated with vehicle navigation;

generating a semantic position estimate associated with an object represented in the semantic map based on semantic data location information associated with the object;

aligning the semantic position estimate with a physical feature that corresponds with the object to provide a final position for the semantic position estimate that is a two-dimensional position;

applying one or more three-dimensional semantic labels to the geometric map based on the final position of the semantic position estimate;

generating a warped semantic map by warping the semantic map based on the one or more three-dimensional semantic labels; and

estimating a position of the object within an image associated with the geometric map by applying the semantic position estimate to the image, wherein aligning the semantic position estimate includes receiving the final position for the semantic position estimate, including the two-dimensional position within the image, and wherein applying the one or more three-dimensional semantic labels comprises translating the final position of the semantic position estimate within the image into a three-dimensional position within the geometric map.

12. The non-transitory computer-readable storage medium of claim 11 , wherein the instructions, when executed by the at least one processor of the computing system, further cause the computing system to perform:

receiving a set of images, including the image, associated with the geometric map.

13. The non-transitory computer-readable storage medium of claim 12 , wherein receiving the final position for the semantic position estimate comprises:

receiving a user input adjusting a position of the semantic position estimate within the image.

Assignments (3)
CHANGE OF NAME Recorded Jul 25, 2023
From: WOVEN PLANET NORTH AMERICA, INC.
To: WOVEN BY TOYOTA, U.S., INC.
Reel/Frame 064371/0565 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 13, 2023
From: LYFT, INC.
To: WOVEN PLANET NORTH AMERICA, INC.
Reel/Frame 063936/0279 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jul 12, 2021
From: LYFT, INC.; MAGNA AUTONOMOUS SYSTEMS, LLC
To: LYFT, INC.; MAGNA AUTONOMOUS SYSTEMS, LLC
Reel/Frame 057434/0623 →