All over the examples I read the phrase "retrieves spatial location coordinates: x,y,z relative to the center of depth map." (latest is spatial_tiny_yolo example)
I want to know the distance of an object from the device (Oak-D-Lite in my case), not the distance from a variable point in the "center of depth map".
I saw reference to computing the x,y,z as the average x, average y, average z in an object's bounding box on the depth map.
Q: Are these x,y,z the distance from the "x,y,z of the center of the whole frame" ?
Q: Am I not understanding the terminology, or do I need to perform some on-host math to get the desired "(average) distance of recognized object from Oak-D-Lite device"?
Q: If I wanted the closest distance to the recognized object instead of the average distance of the depth map in the bounding box, can I do that in the Oak-D-Lite, or only on host as a post process?