Three-dimensional segmentation annotation

Patent metadata

Field	Value
Publication number	US-11928824-B2
Application number	US-202117472849-A
Country	US
Kind code	B2
Filing date	Sep 13, 2021
Priority date	Sep 13, 2021
Publication date	Mar 12, 2024
Grant date	Mar 12, 2024

Abstract

An approach is provided in which the approach receives an image that includes multiple image points and constructs a plane in the image based on a first subset of the plurality of image points. The approach identifies a second subset of the image points that belong to the plane and are not part of the first subset of image points, and removes the first subset of image points and the second subset of image points form the image points. The approach annotates the remaining subset of image points in the image.

First claim

The invention claimed is:

1. A computer-implemented method comprising:
receiving an image comprising a plurality of image points;
constructing a first plane in the image based on a first subset of the plurality of image points that belong to the first plane;
identifying a second subset of the plurality of image points that belong to the first plane and are not part of the first subset of image points;
removing the first subset of image points and the second subset of image points from the plurality of image points, wherein removing produces a remaining subset of the plurality of image points in the image; and
annotating the remaining subset of image points.

2. The computer-implemented method of claim 1 wherein the first subset of image points only comprises a first image point, a second image point, and a third image point.

3. The computer-implemented method of claim 1 further comprising:
calculating a plurality of distances between the first plane and the second subset of image points; and
adding the second subset of image points to the first plane in response to determining that their corresponding distances to the first plane are below a threshold.

4. The computer-implemented method of claim 1 further comprising:
performing a density analysis on the second subset of image points; and
removing one or more of the subset of image points from the first plane based on the density analysis and a density threshold.

5. The computer-implemented method of claim 1 further comprising:
constructing a second plane in the image based on a third subset of image points belonging to the second plane and included in the remaining subset of image points;
identifying a fourth subset of image points from the remaining subset of image points that belong to the second plane and are not part of the third subset of image points;
removing the third subset of image points and the fourth subset of image points from the remaining subset of image points to produce a final remaining subset of image points; and
annotating the final remaining subset of image points.

6. The computer-implemented method of claim 5 further comprising:
performing a 3D cluster analysis on the final remaining subset of image points; and
providing a set of label candidates of the final remaining subset of image points in response to performing the 3D cluster analysis.

7. The computer-implemented method of claim 1 wherein the image is a point cloud image, and wherein the first plane corresponds to a flat surface in the point cloud image selected from the group consisting of a wall, a floor, and a ground.

8. An information handling system comprising:
one or more processors;
a memory coupled to at least one of the processors;
a set of computer program instructions stored in the memory and executed by at least one of the processors in order to perform actions of:
receiving an image comprising a plurality of image points;
constructing a first plane in the image based on a first subset of the plurality of image points that belong to the first plane;
identifying a second subset of the plurality of image points that belong to the first plane and are not part of the first subset of image points;
removing the first subset of image points and the second subset of image points from the plurality of image points, wherein removing produces a remaining subset of the plurality of image points in the image; and
annotating the remaining subset of image points.

9. The information handling system of claim 8 wherein the first subset of image points only comprises a first image point, a second image point, and a third image point.

10. The information handling system of claim 8 wherein the processors perform additional actions comprising:
calculating a plurality of distances between the first plane and the second subset of image points; and
adding the second subset of image points to the first plane in response to determining that their corresponding distances to the first plane are below a threshold.

11. The information handling system of claim 8 wherein the processors perform additional actions comprising:
performing a density analysis on the second subset of image points; and
removing one or more of the subset of image points from the first plane based on the density analysis and a density threshold.

12. The information handling system of claim 8 wherein the processors perform additional actions comprising:
constructing a second plane in the image based on a third subset of image points belonging to the second plane and included in the remaining subset of image points;
identifying a fourth subset of image points from the remaining subset of image points that belong to the second plane and are not part of the third subset of image points;
removing the third subset of image points and the fourth subset of image points from the remaining subset of image points to produce a final remaining subset of image points; and
annotating the final remaining subset of image points.

13. The information handling system of claim 12 wherein the processors perform additional actions comprising:
performing a 3D cluster analysis on the final remaining subset of image points; and
providing a set of label candidates of the final remaining subset of image points in response to performing the 3D cluster analysis.

14. The information handling system of claim 8 wherein the image is a point cloud image, and wherein the first plane corresponds to a flat surface in the point cloud image selected from the group consisting of a wall, a floor, and a ground.

15. A computer program product stored in a computer readable storage medium, comprising computer program code that, when executed by an information handling system, causes the information handling system to perform actions comprising:
receiving an image comprising a plurality of image points;
constructing a first plane in the image based on a first subset of the plurality of image points that belong to the first plane;
identifying a second subset of the plurality of image points that belong to the first plane and are not part of the first subset of image points;
removing the first subset of image points and the second subset of image points from the plurality of image points, wherein removing produces a remaining subset of the plurality of image points in the image; and
annotating the remaining subset of image points.

16. The computer program product of claim 15 wherein the first subset of image points only comprises a first image point, a second image point, and a third image point.

17. The computer program product of claim 15 wherein the information handling system performs further actions comprising:
calculating a plurality of distances between the first plane and the second subset of image points; and
adding the second subset of image points to the first plane in response to determining that their corresponding distances to the first plane are below a threshold.

18. The computer program product of claim 15 wherein the information handling system performs further actions comprising:
performing a density analysis on the second subset of image points; and
removing one or more of the subset of image points from the first plane based on the density analysis and a density threshold.

19. The computer program product of claim 15 wherein the information handling system performs further actions comprising:
constructing a second plane in the image based on a third subset of image points belonging to the second plan

Assignees

• IBM

Inventors

• Liu Xue Ping
• Zhang Dan
• Ding Yuan Yuan
• Xin Chao
• Li Fan
• Zhang Hong Bing
• Min Xu

Classifications

• G06T7/136 Primary

  involving thresholding · CPC title

• G06F18/23

  Clustering techniques · CPC title

• G06T2200/04

  involving 3D image data · CPC title

• G06T2207/10028

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

• G06T7/11 Primary

  Region-based segmentation · CPC title