The answer was not helpful…
before you can ask about efficiency, you have to have something functional first. I don’t see that.
what I see is that you copied the code from his blog and tried to modify it. does that code even run? I’d bet it does not run, it throws exceptions. at one point you take the 2D points of a bounding box and act like these points had three coordinates.
please just get a proper course on image processing and computer vision. do not get anything from that dude. all I see him offering is a bunch of “recipes” implemented in code. that’s not teaching you the approaches or giving you a proper broad overview of the field.
yes. depends on what you need it for, and what type of camera you have, and what object it is.