1
$\begingroup$

I am new to the field of Camera Calibration. We have successfully calibrated our camera through classic method (using checkerboard etc). I recently realized there is a field of research for Automatic Camera Calibration. I expect these methods to only be capable of estimation intrinsic camera parameters. However, from my early researches, it seems to me, that they can estimation extrinsic camera parameters (X,Y,Z position in the world coordinates). It occurred to me that:

  • If these methods does not need any control points (real world points with measured distances)?

    If it is so, how they do that, I mean how they know where is the camera in the world?

$\endgroup$
  • 1
    $\begingroup$ Can I please ask you to clarify which algorithms you have reviewed so far? In the meantime, I suppose that you are referring to something like SLAM. In that case (and derivatives of) location is inferred relative to some starting point (e.g. 0,0,0). Through SLAM, you could attempt to recognise geographical features as well and after some processing, guess what your most likely GPS fix might be. $\endgroup$ – A_A Aug 24 '16 at 5:33
  • $\begingroup$ Thanks for suggestion, about the auto-correlation I've just heared a few algorithm names and about classical approach I am using the method "Recovering the camera parameters from a transformation matrix,"TM Strat - Readings in Computer Vision, 1987 . $\endgroup$ – MimSaad Aug 24 '16 at 10:55

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.