We purchased the 3d camera to evaluate its suitability for our augmented reality system, which currently uses 4 cameras to track head position, position of a half transparent mirror and a light pen position. We would like to extend our system with hand tracking.
To use the 3d camera in our system, the camera has to be registered with the rest of the system. Intristic camera parameters need to be known with a good precision. I suppose those are set with PROPERTY_DEPTH_FOCAL_LENGTH and PROPERTY_DEPTH_PRINCIPAL_POINT parameters for the depth camera. I am not sure though what PROPERTY_DEPTH_FIELD_OF_VIEW and PROPERTY_DEPTH_SENSOR_RANGE mean. Maybe an illustration would enlighten the parameter meaning?
The camera information tool shows camera calibration parameters including parameters to correct the lens distortion. Are those parameters measured for each camera device or are those parameters fixed for all cameras manufactured? Where are the parameters stored in the system, what is their meaning and how are they used by the SDK? Is it meaningful for us to calibrate the camera to improve registration in our augmented reality system?
What is the depth resolution of the camera? The depth values seem to be very noisy. When averaging a single pixel 30x, I believe the resolution is around 3mm at the further distances. What exactly does the SDK when the PROPERTY_DEPTH_SMOOTHING is enabled?