Parallel data capture system

1. A server, comprising:

a communications interface connecting the server with a plurality of capture nodes, each capture node including (i) a plurality of cameras having overlapping fields of view directed towards a capture volume having an object present therein, and (ii) at least one computing device configured to obtain two-dimensional images via the cameras and generate an initial point cloud from the two-dimensional images depicting at least a portion of the capture volume; and

a processor connected with the communications interface, the processor configured to:

receive, from the at least one computing device: (i) the initial point cloud, and (ii) boundary values corresponding to the initial point cloud;

generate a bounding box from the boundary values received from the at least one computing device;

select respective portions of each initial point cloud based on the bounding box; and

combine the selected portions of each initial point cloud to generate a combined point cloud.

2. The server of claim 1 , further comprising a memory; wherein the processor is further configured to:

retrieve calibration data from the memory; and

register the initial point clouds to a common frame of reference according to the calibration data.

3. The server of claim 1 , wherein the boundary values include:

at least a first value defining a minimum extent in a first dimension;

at least a second value defining a maximum extent in the first dimension;

at least a third value defining a minimum extent in a second dimension; and

at least a fourth value defining a maximum extent in the second dimension.

4. The server of claim 3 , wherein the first and second dimensions are horizontal.

5. The server of claim 4 , wherein the processor is further configured, in order to generate the bounding box, to:

generate a first vertical surface based on the first value;

generate a second vertical surface based on the second value;

generate a third vertical surface based on the third value; and

generate a fourth vertical surface based on the fourth value.

6. The server of claim 3 , wherein the boundary values received from the capture nodes include a plurality of minimum boundary values defining minimum extents in the first dimension; and

wherein the processor is further configured to select the first value from the plurality of values.

7. The server of claim 1 , wherein the processor is further configured, in order to select the respective portions of each initial point cloud, to:

determine a center of the bounding box;

generate a set of cutting planes based on the center; and

for each initial point cloud, select the respective portion based on the bounding box and a subset of the cutting planes.

8. The server of claim 1 , wherein the processor is further configured, prior to receiving the initial point clouds, to detect the object present in the capture volume and generate a trigger command for transmission to the capture nodes.

9. The server of claim 1 , wherein the processor is further configured to determine, based on the combined point cloud, a dimension of the object present in the capture volume.

10. A method, comprising:

at a server, connecting with a plurality of capture nodes, each capture node including (i) a plurality of cameras having overlapping fields of view directed towards a capture volume having an object present therein, and (ii) at least one computing device configured to obtain two-dimensional images via the cameras and generate an initial point cloud from the two-dimensional images depicting at least a portion of the capture volume;

receiving at the server, from the at least one computing device: (i) the initial point cloud, and (ii) boundary values corresponding to the initial point cloud;

generating a bounding box from the boundary values received from the at least one computing device;

selecting respective portions of each initial point cloud based on the bounding box; and

combining the selected portions of each initial point cloud to generate a combined point cloud.

11. The method of claim 10 , further comprising:

retrieving calibration data; and

registering the initial point clouds to a common frame of reference according to the calibration data.

12. The method of claim 10 , wherein the boundary values include:

at least a first value defining a minimum extent in a first dimension;

at least a second value defining a maximum extent in the first dimension;

at least a third value defining a minimum extent in a second dimension; and

at least a fourth value defining a maximum extent in the second dimension.

13. The method of claim 12 , wherein the first and second dimensions are horizontal.

14. The method of claim 13 , wherein generating the bounding box comprises:

generating a first vertical surface based on the first value;

generating a second vertical surface based on the second value;

generating a third vertical surface based on the third value; and

generating a fourth vertical surface based on the fourth value.

15. The method of claim 12 , wherein

the boundary values received from the capture nodes include a plurality of minimum boundary values defining minimum extents in the first dimension;

the method further comprises selecting the first value from the plurality of values.

16. The method of claim 10 , wherein selecting the respective portions of each initial point cloud comprises:

determining a center of the bounding box;

generating a set of cutting planes based on the center; and

for each initial point cloud, selecting the respective portion based on the bounding box and a subset of the cutting planes.

17. The method of claim 10 , further comprising, prior to receiving the initial point clouds,

detecting the object present in the capture volume and generating a trigger command for transmission to the capture nodes.

18. The method of claim 10 , further comprising: determining, based on the combined point cloud, a dimension of the object present in the capture volume.