4
$\begingroup$

AprilTag and Aruco are both popular methods for vision-based fiducial marker pose estimation.

What are the pros and cons of each fiducial marker pose estimation systems?

$\endgroup$
1
5
$\begingroup$

Aruco (as implemented in OpenCV)

  • pros
    • Easy to set up (with readily available aruco marker generator, opencv & ros implementation, etc.)
    • fewer false detection (with default parameters)
  • cons
    • Newer versions of aruco is GPL licensed, hence opencv is stuck on an old implementation of aruco when it was still BSD.
    • More susceptible to rotational ambiguity at medium to long ranges
    • More tuning parameters
    • More computationally intensive

AprilTag (as implemented in apriltag_ros)

  • pros
    • BSD license
    • Fewer tuning parameters
    • Works fairly well even at long range
    • Used by NASA
    • More flexible marker design (e.g. markers not necessarily square)
    • Less computationally intensive
  • cons
    • less straight forward to setup (no opencv implementation AFAIK, only ros implementation, slightly more steps to obtain markers)
    • more false detection (with default parameters)
$\endgroup$
2
  • 1
    $\begingroup$ Why "More susceptible to rotational ambiguity" for Aruco? The document you link explains that the rotational ambiguity problem exists for any pose estimation of coplanar points. Given that an AprilTag is also flat, why would it suffer less from this problem? Thanks! $\endgroup$ – nh2 Dec 30 '20 at 19:19
  • 2
    $\begingroup$ I'm not sure why, the observation was based on experimental evidence. For marker of the same size and same distance away, rotational ambiguity occurred less frequently for AprilTag $\endgroup$ – Rufus Dec 31 '20 at 3:28

Not the answer you're looking for? Browse other questions tagged or ask your own question.