Analysis of point cloud data using depth maps

Patent metadata

Field	Value
Publication number	US-11301697-B2
Application number	US-202016938473-A
Country	US
Kind code	B2
Filing date	Jul 24, 2020
Priority date	Feb 27, 2018
Publication date	Apr 12, 2022
Grant date	Apr 12, 2022

Abstract

Various types of systems or technologies can be used to collect data in a 3D space. For example, LiDAR (light detection and ranging) and RADAR (radio detection and ranging) systems are commonly used to generate point cloud data for 3D space around vehicles, for such functions as localization, mapping, and tracking. This disclosure provides improved techniques for processing the point cloud data that has been collected. The improved techniques include mapping 3D point cloud data points into a 2D depth map, fetching a group of the mapped 3D point cloud data points that are within a bounded window of the 2D depth map; and generating geometric space parameters based on the group of the mapped 3D point cloud data points. The generated geometric space parameters may be used for object motion, obstacle detection, freespace detection, and/or landmark detection for an area surrounding a vehicle.

First claim

What is claimed is:

1. A method of analyzing data points in a point cloud system, comprising:
receiving three dimensional (3D) point cloud data points;
mapping the received 3D point cloud data points into a 2D depth map;
fetching a group of the mapped 3D point cloud data points that are within a bounded window of the 2D depth map; and
generating, using a processor, geometric space parameters based on the group of the mapped 3D point cloud data points.

2. The method as recited in claim 1 , further comprising:
processing the group of the mapped 3D point cloud data points in parallel.

3. The method as recited in claim 1 , further comprising:
utilizing a LiDAR system to generate, at a periodic interval, slices of the received 3D point cloud data points; and
wherein the received 3D point cloud data points represent part of a geometric space of a vehicle.

4. The method as recited in claim 3 , further comprising:
varying a size of the bounded window of the 2D depth map based an angular separation between the slices.

5. The method as recited in claim 3 , wherein the vehicle is an autonomous or semi-autonomous vehicle.

6. The method as recited in claim 3 , further comprising:
determining a trajectory of an object and a trajectory of the vehicle; and
using the trajectories to determine if the vehicle needs to slow down or turn to prevent the object from coming into contact with the vehicle.

7. The method as recited in claim 3 , further comprising:
storing each of the slices as a portion of the 2D depth map; and
utilizing, for each of the slices, a first angle parameter for a horizontal dimension of the each slice, and a second angle parameter for a vertical dimension of the each slice, wherein the first angle parameter is equivalent for each of the slices analyzed and the second angle parameter is equivalent for each of the slices analyzed.

8. The method as recited in claim 1 , wherein the fetching includes fetching all of the mapped 3D point cloud data points in the group.

9. The method as recited in claim 1 , wherein the fetching includes fetching some of the mapped 3D point cloud data points that are in the group.

10. The method as recited in claim 1 , wherein the group of the mapped 3D point cloud data points is processed in parallel using a graphics processing unit (GPU), and the processor includes the GPU.

11. The method as recited in claim 1 , wherein the mapping includes translating depth information of the received 3D point cloud data points to corresponding data points in the 2D depth map.

12. The method as recited in claim 11 , wherein the translating includes using a minimum parameter calculation.

13. The method as recited in claim 11 , wherein the translating includes using an average parameter calculation.

14. The method as recited in claim 1 , wherein a separate GPU processing thread is spawned for each of the mapped 3D point cloud data points in the group.

15. The method as recited in claim 1 , wherein the geometric space parameters are generated for at least one of obstacle detection, freespace detection, or landmark detection.

16. A point cloud data analysis system to generate geometric space parameters, comprising:
a receiver operable to receive three dimensional (3D) point cloud data points representative of part of a geometric space surrounding a vehicle; and
a graphics processing unit (GPU) operable to map the received 3D point cloud data points into a two dimensional (2D) depth map, to fetch a group of the mapped 3D point cloud data points that are within a bounded window of the 2D depth map, and to generate parameters of the geometric space based on the group of the mapped 3D point cloud data points.

17. The system as recited in claim 16 , further comprising:
an initiator, operable to initiate a generation of the 3D point cloud data points at a periodic interval.

18. The system as recited in claim 17 , wherein the initiator is a LiDAR system.

19. The system as recited in claim 16 , wherein the vehicle is one of an autonomous vehicle and a semi-autonomous vehicle.

20. The system as recited in claim 16 , wherein the parameters are associated with at least one of a moving object, obstacle detection, freespace detection, or landmark detection.

21. A computer program product having a series of operating instructions stored on a non-transitory computer-readable medium that directs a data processing apparatus, when executed thereby, to perform operations comprising:
receiving three dimensional (3D) point cloud data points, at a periodic interval, representative of part of a geometric space surrounding a vehicle;
mapping the received 3D point cloud data points into a two dimensional (2D) depth map;
fetching a group of the mapped 3D point cloud data points that are within a bounded window of the 2D depth map; and
generating parameters of the geometric space based on the group of the mapped 3D point cloud data points.

22. The computer program product as recited in claim 21 , wherein the operations further comprise processing the group of the mapped 3D point cloud data points in parallel utilizing a graphics processing unit (GPU).

23. The computer program product as recited in claim 21 , wherein the mapping includes translating depth information of the received 3D point cloud data points to corresponding data points in the 2D depth map.

24. The computer program product as recited in claim 21 , wherein the fetching includes fetching all of the mapped 3D point cloud data points in the group.

25. The computer program product as recited in claim 21 , wherein the parameters for generated for at least one of obstacle detection, freespace detection, or landmark detection.

Assignees

• Nvidia Corp

Inventors

• Kulkarni Ishwar
• Eden Ibrahim
• Kroepfl Michael
• Nister David

Classifications

• G06T11/10

  Texturing; Colouring; Generation of textures or colours (retouching, inpainting or scratch removal G06T5/77) · CPC title

• G06T2207/10028

  Range image; Depth image; 3D point clouds · CPC title

• G06T2207/30241

  Trajectory · CPC title

• G01S17/931

  of land vehicles · CPC title

• G01S17/89

  for mapping or imaging · CPC title