I'm trying to get the best possible pose estimate. My big question is in the title - is it better to have, say, a 2x2 Arucoboard, or a single marker with 4x the area (of those in the Arucoboard)? Occlusion should not be a problem for my application.
Also, I already have seen a large improvement by using corner refinement in the Aruco detector parameters (off by default). What other detector settings should I adjust to improve the pose estimate? How about camera settings? Does having a well-calibrated camera make a big difference for accurate pose estimation?