Hi I am pretty new to computer vision and currently experimenting with an OAK-S S2 camera my company has just gotten hold of.
I am trying to understand what the point cloud data actually is. What I want is to get a series of x,y and z co-ordinates of what the camera can see.
I am working off this example and the linked line is the important one
https://github.com/luxonis/depthai-experiments/blob/master/gen2-pointcloud/device-pointcloud/main.py#L175
In this case pcl_data is always 256000 records, each of them a 3 element array. Most of the arrays are [0, 0, 0] in value.
But some have values like
[ 0.21325 1.544 3.056 ]
[ 0.227 1.586 3.138 ]
[ 0.24475 1.652 3.268 ]
What do these values mean?
My ultimate goal is not to just recognise an object and how far away it is but for a well known shape be able to work out it's orientation in 3d space at to within a few degrees. Measuring points on the edge these objects in 3d seemed the best way to do this.