Please excuse me if this is not the proper place to ask this. I just need some general advice.

I have set up a very basic homemade motion capture system where I put three TrackIR cameras from NaturalPoint in a three fixed positions. I want to capture movement of a single object in three axis. So I laid out two of the cameras on both sides of my working area, plus the third camera above the object, pointing downwards.

I then use reflective markers to track the position of my object in the cameras.

I don't do any calculations on my system aside from a simple coordinate transformation from that of the real world to that of my virtual world. I do this separately, for each of the cameras. Since they are in fixed positions I already know how movements translate from one "world" to the other.

Is there any way that I can improve my mo-cap system? Should I apply some kind of triangulation (isn't triangulation implicit in my system already, given that Im using fixed positions for my cameras)? What filtering could I use to account for punctual lost of tracking elements?

Note that I want (ie. need) to stick to my current hardware.

  • $\begingroup$ You don't really say what's wrong with the current system. Is there anything that doesn't work? $\endgroup$ – Peter K. Sep 1 '13 at 0:58
  • $\begingroup$ It does work but it doesn't keep temporal or spatial correspondence with markers that well. I was thinkning maybe I could improve it somehow. $\endgroup$ – MichelReap Sep 1 '13 at 7:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.