Point cloud registration for lidar labeling
The subject disclosure relates to techniques for detecting an object. A process of the disclosed technology can include steps for receiving three-dimensional (3D) Light Detection and Ranging (LiDAR) data of the object at a first time, generating a first point cloud based on the 3D LiDAR data at the first time, receiving 3D LiDAR data of the object at a second time, generating a second point cloud based on the 3D LiDAR data at the second time, aggregating the first point cloud and the second point cloud to form an aggregated point cloud, and placing a bounding box around the aggregated point cloud. Systems and machine-readable media are also provided.
1. A computer-implemented method for detecting an object using Light Detection and Ranging (LiDAR), the method comprising:
receiving three-dimensional (3D) Light Detection and Ranging (LiDAR) data of the object at a first time;
identifying, using a mixture model on the 3D LiDAR data, that a first set of points of the 3D LiDAR data represent a first portion of the object;
generating a first 3D point cloud based on the 3D LiDAR data at the first time, wherein the first 3D point cloud is associated with the first portion of the object;
receiving additional 3D LiDAR data of the object at a second time;
identifying, using the mixture model on the additional 3D LiDAR data, that a second set of points of the additional 3D LiDAR data represent a second portion of the object;
generating a second 3D point cloud based on the additional 3D LiDAR data at the second time, wherein the second 3D point cloud is associated with the second portion of the object;
aggregating the first 3D point cloud and the second 3D point cloud to form an aggregated 3D point cloud associated with the first portion of the object and the second portion of the object;
placing a bounding box around the aggregated 3D point cloud; and
tracking the object based on the bounding box.
2. The computer-implemented method of claim 1 , wherein the first 3D point cloud is based on the first set of points of the 3D LiDAR data at the first time.
3. The computer-implemented method of claim 1 , wherein the second 3D point cloud is based on the second set of points of the additional 3D LiDAR data at the second time.
4. The computer-implemented method of claim 1 , wherein the first set of points of the 3D LiDAR data is a subset of points representing an environment surrounding the LiDAR.
5. The computer-implemented method of claim 1 , wherein the LiDAR data is indicative of at least one of points in 3D space, speed, and direction.
6. The computer-implemented method of claim 1 , further comprising:
labelling the bounding box; and
continuously identifying the object being tracked.
7. The computer-implemented method of claim 1 , wherein the bounding box is selectable to cause an alteration to the bounding box.
8. A system for detecting an object using Light Detection and Ranging (LiDAR), the system comprising:
one or more processors; and
a computer-readable medium comprising instructions stored therein, which when executed by the processors, cause the processors to perform operations comprising:
receiving three-dimensional (3D) Light Detection and Ranging (LiDAR) data of the object at a first time;
identifying, using a mixture model on the 3D LiDAR data, that a first set of points of the 3D LiDAR data represent a first portion of the object;
generating a first 3D point cloud based on the 3D LiDAR data at the first time, wherein the first 3D point cloud is associated with the first portion of the object;
receiving additional 3D LiDAR data of the object at a second time;
identifying, using the mixture model on the additional 3D LiDAR data, that a second set of points of the additional 3D LiDAR data represent a second portion of the object;
generating a second 3D point cloud based on the additional 3D LiDAR data at the second time, wherein the second 3D point cloud is associated with the second portion of the object;
aggregating the first 3D point cloud and the second 3D point cloud to form an aggregated 3D point cloud associated with the first portion of the object and the second portion of the object;
placing a bounding box around the aggregated 3D point cloud; and
tracking the object based on the bounding box.
9. The system of claim 8 , wherein the first 3D point cloud is based on the first set of points of the 3D LiDAR data at the first time.
10. The system of claim 8 , wherein the second 3D point cloud is based on the second set of points of the additional 3D LiDAR data at the second time.
11. The system of claim 8 , wherein the first set of points of the 3D LiDAR data is a subset of points representing an environment surrounding the LiDAR.
12. The system of claim 8 , wherein the LiDAR data is indicative of at least one of points in 3D space, speed, and direction.
13. The system of claim 8 , wherein the instructions, when executed by the processors, cause the processors to further perform operations comprising:
labelling the bounding box; and
continuously identifying the object being tracked.
14. The system of claim 8 , wherein the bounding box is selectable to cause an alteration to the bounding box.
15. A non-transitory computer-readable storage medium comprising instructions stored therein, which when executed by one or more processors, cause the processors to perform operations for detecting an object using Light Detection and Ranging, the operations comprising:
receiving three-dimensional (3D) Light Detection and Ranging (LiDAR) data of the object at a first time;
identifying, using a mixture model on the 3D LiDAR data, that a first set of points of the 3D LiDAR data represent a first portion of the object;
generating a first 3D point cloud based on the 3D LiDAR data at the first time, wherein the first 3D point cloud is associated with the first portion of the object;
receiving additional 3D LiDAR data of the object at a second time;
identifying, using the mixture model on the additional 3D LiDAR data, that a second set of points of the additional 3D LiDAR data represent a second portion of the object;
generating a second 3D point cloud based on the additional 3D LiDAR data at the second time, wherein the second 3D point cloud is associated with the second portion of the object;
aggregating the first 3D point cloud and the second 3D point cloud to form an aggregated 3D point cloud associated with the first portion of the object and the second portion of the object;
placing a bounding box around the aggregated 3D point cloud; and
tracking the object based on the bounding box.
16. The non-transitory computer-readable storage medium of claim 15 , wherein the first 3D point cloud is based on the first set of points of the 3D LiDAR data at the first time.
17. The non-transitory computer-readable storage medium of claim 15 , wherein the second 3D point cloud is based on the second set of points of the additional 3D LiDAR data at the second time.
18. The non-transitory computer-readable storage medium of claim 15 , wherein the first set of points of the 3D LiDAR data is a subset of points representing an environment surrounding the LiDAR.
19. The non-transitory computer-readable storage medium of claim 15 , wherein the LiDAR data is indicative of at least one of points in 3D space, speed, and direction.
20. The non-transitory computer-readable storage medium of claim 15 , wherein the bounding box is selectable to cause an alteration to the bounding box.