Torsten Sattler
Yes, this will work.
If you just use a single pair, your scaling factor will be affected by noise in the individual pose estimates (both arcore and Colmap produce noisy poses). It would be...
Have you tried the model_aligner?
You can convert binary files to different formats, including text files, using the `model_converter`.
This seems to be an issue of the script from the LLFF repository, and not a problem generated by Colmap. Did you ask for help in the LLFF github repository?
Yes, the `refine_focal_length` parameter affects all cameras. It should be fairly easy to modify the source code though as to only refine certain cameras.
I don't think that there is an easy fix for this. Rather, you would probably need to modify source code, e.g., you could try to look at the distribution of...
I am not sure I understand.
I understand the basic idea, but I am not sure how you would measure the overall error when you say "we can try to remove and remap some of them...
1. This is not necessarily correct: at the true pose, you might have more 2D-3D correspondences than at the wrong pose. At the same time, you already have quite low...