Granted Patent B2

US 12,367,542 · App. 17/884,684 · Granted Jul 22, 2025

Inventors: Zhisong Liu (Shenzhen, CN); Zijia Wang (WeiFang, CN); Zhen Jia (Shanghai, CN)

Assignee: Dell Products L.P.

G06T3/40G06T17/00G06T2210/36G06T2210/56

Abstract

Claims (85)

1. A point cloud processing method, comprising:

obtaining a first point cloud based on an input point cloud, a point number of the first point cloud being greater than a point number of the input point cloud;

obtaining a first group of point clouds based on the first point cloud, the first group of point clouds comprising a plurality of point clouds;

obtaining a second group of point clouds based on the input point cloud and the first group of point clouds, the second group of point clouds comprising a plurality of point clouds; and

obtaining a target point cloud based on the first point cloud and the second group of point clouds, a point number of the target point cloud being greater than the point number of the input point cloud;

wherein obtaining the second group of point clouds based on the input point cloud and the first group of point clouds comprises:

obtaining at least one additional group of point clouds based on the input point cloud and the first group of point clouds, the at least one additional group of point clouds comprising a plurality of point clouds; and

obtaining the second group of point clouds based on the at least one additional group of point clouds;

wherein the at least one additional group of point clouds comprises a third group of point clouds, a point number of the second group of point clouds being greater than a point number of the third group of point clouds.

2. The method according to claim 1 , wherein obtaining the first point cloud based on the input point cloud comprises:

generating a feature value of the input point cloud based on coordinates of the input point cloud;

obtaining a feature value of the first point cloud based on the feature value of the input point cloud; and

generating coordinates of the first point cloud based on the feature value of the first point cloud.

3. The method according to claim 2 , wherein obtaining the feature value of the first point cloud based on the feature value of the input point cloud comprises:

upsampling the feature value of the input point cloud to obtain a first feature value;

downsampling the first feature value to obtain a second feature value;

calculating a difference value between the feature value of the input point cloud and the second feature value to obtain a first residual feature value;

upsampling the first residual feature value to obtain a third feature value; and

adding the third feature value and the first feature value to obtain the feature value of the first point cloud.

4. The method according to claim 1 , wherein obtaining the first group of point clouds based on the first point cloud comprises:

downsampling the first point cloud to obtain the first group of point clouds.

5. The method according to claim 1 , wherein obtaining the target point cloud based on the first point cloud and the second group of point clouds comprises:

performing accumulation calculation on coordinates of the first point cloud and coordinates of each of corresponding point clouds in the second group of point clouds to obtain the target point cloud.

6. The method according to claim 1 , wherein obtaining the third group of point clouds based on the input point cloud and the first group of point clouds comprises:

calculating a difference value between coordinates of the input point cloud and coordinates of each of corresponding point clouds in the first group of point clouds respectively to obtain a group of residuals.

7. The method according to claim 1 , wherein obtaining the second group of point clouds based on the third group of point clouds comprises:

generating a feature value of each point cloud in the third group of point clouds based on coordinates of each point cloud in the third group of point clouds;

obtaining a feature value of each point cloud in the second group of point clouds based on the feature value of each point cloud in the third group of point clouds; and

generating coordinates of each point cloud in the second group of point clouds based on the feature value of each point cloud in the second group of point clouds.

8. The method according to claim 7 , wherein obtaining the feature value of each point cloud in the second group of point clouds based on the feature value of each point cloud in the third group of point clouds comprises:

upsampling the feature value of each point cloud in the third group of point clouds to obtain a fourth feature value;

downsampling the fourth feature value to obtain a fifth feature value;

calculating a difference value between the feature value of each point cloud in the third group of point clouds and the fifth feature value to obtain a second residual feature value;

upsampling the second residual feature value to obtain a sixth feature value; and

adding the sixth feature value and the fourth feature value to obtain the feature value of each point cloud in the second group of point clouds.

9. An electronic device for point cloud processing, comprising:

at least one processor; and

memory coupled to the at least one processor and having instructions stored therein, wherein the instructions, when executed by the at least one processor, cause the electronic device to perform actions comprising:

obtaining a first point cloud based on an input point cloud, a point number of the first point cloud being greater than a point number of the input point cloud;

obtaining a first group of point clouds based on the first point cloud, the first group of point clouds comprising a plurality of point clouds;

obtaining a second group of point clouds based on the input point cloud and the first group of point clouds, the second group of point clouds comprising a plurality of point clouds; and

obtaining a target point cloud based on the first point cloud and the second group of point clouds, a point number of the target point cloud being greater than the point number of the input point cloud;

wherein obtaining the second group of point clouds based on the input point cloud and the first group of point clouds comprises:

obtaining at least one additional group of point clouds based on the input point cloud and the first group of point clouds, the at least one additional group of point clouds comprising a plurality of point clouds; and

obtaining the second group of point clouds based on the at least one additional group of point clouds;

wherein the at least one additional group of point clouds comprises a third group of point clouds, a point number of the second group of point clouds being greater than a point number of the third group of point clouds.

10. The electronic device according to claim 9 , wherein obtaining the first point cloud based on the input point cloud comprises:

generating a feature value of the input point cloud based on coordinates of the input point cloud;

obtaining a feature value of the first point cloud based on the feature value of the input point cloud; and

generating coordinates of the first point cloud based on the feature value of the first point cloud.

11. The electronic device according to claim 10 , wherein obtaining the feature value of the first point cloud based on the feature value of the input point cloud comprises:

upsampling the feature value of the input point cloud to obtain a first feature value;

downsampling the first feature value to obtain a second feature value;

calculating a difference value between the feature value of the input point cloud and the second feature value to obtain a first residual feature value;

upsampling the first residual feature value to obtain a third feature value; and

adding the third feature value and the first feature value to obtain the feature value of the first point cloud.

12. The electronic device according to claim 9 , wherein obtaining the first group of point clouds based on the first point cloud comprises:

downsampling the first point cloud to obtain the first group of point clouds.

13. The electronic device according to claim 9 , wherein obtaining the target point cloud based on the first point cloud and the second group of point clouds comprises:

performing accumulation calculation on coordinates of the first point cloud and coordinates of each of corresponding point clouds in the second group of point clouds to obtain the target point cloud.

14. The electronic device according to claim 9 , wherein obtaining the third group of point clouds based on the input point cloud and the first group of point clouds comprises:

calculating a difference value between coordinates of the input point cloud and coordinates of each of corresponding point clouds in the first group of point clouds respectively to obtain a group of residuals.

15. The electronic device according to claim 9 , wherein obtaining the second group of point clouds based on the third group of point clouds comprises:

generating a feature value of each point cloud in the third group of point clouds based on coordinates of each point cloud in the third group of point clouds;

obtaining a feature value of each point cloud in the second group of point clouds based on the feature value of each point cloud in the third group of point clouds; and

generating coordinates of each point cloud in the second group of point clouds based on the feature value of each point cloud in the second group of point clouds.

16. The electronic device according to claim 15 , wherein obtaining the feature value of each point cloud in the second group of point clouds based on the feature value of each point cloud in the third group of point clouds comprises:

upsampling the feature value of each point cloud in the third group of point clouds to obtain a fourth feature value;

downsampling the fourth feature value to obtain a fifth feature value;

calculating a difference value between the feature value of each point cloud in the third group of point clouds and the fifth feature value to obtain a second residual feature value;

upsampling the second residual feature value to obtain a sixth feature value; and

adding the sixth feature value and the fourth feature value to obtain the feature value of each point cloud in the second group of point clouds.

17. A computer program product tangibly stored on a non-transitory computer-readable medium and comprising machine-executable instructions, wherein the machine-executable instructions, when executed by a machine, cause the machine to perform actions comprising:

obtaining a first point cloud based on an input point cloud, a point number of the first point cloud being greater than a point number of the input point cloud;

obtaining a first group of point clouds based on the first point cloud, the first group of point clouds comprising a plurality of point clouds;

obtaining a second group of point clouds based on the input point cloud and the first group of point clouds, the second group of point clouds comprising a plurality of point clouds; and

obtaining a target point cloud based on the first point cloud and the second group of point clouds, a point number of the target point cloud being greater than the point number of the input point cloud;

wherein obtaining the second group of point clouds based on the input point cloud and the first group of point clouds comprises:

obtaining at least one additional group of point clouds based on the input point cloud and the first group of point clouds, the at least one additional group of point clouds comprising a plurality of point clouds; and

obtaining the second group of point clouds based on the at least one additional group of point clouds;

wherein the at least one additional group of point clouds comprises a third group of point clouds, a point number of the second group of point clouds being greater than a point number of the third group of point clouds.

18. The computer program product according to claim 17 , wherein obtaining the first point cloud based on the input point cloud comprises:

generating a feature value of the input point cloud based on coordinates of the input point cloud;

obtaining a feature value of the first point cloud based on the feature value of the input point cloud; and

generating coordinates of the first point cloud based on the feature value of the first point cloud.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (138)