Torsten Sattler

Can you be more specific about what you are doing? Which parts of Colmap are you running?

Semantic information is currently not supported. Once you have a model, you can project each 3D points into the images from which it was created, look up the corresponding semantic...

Colmap can scale models based on known coordinates of cameras via the `model_aligner`.

Have you had a look here: https://colmap.github.io/faq.html#geo-registration or the documentation provided in the source code: https://github.com/colmap/colmap/blob/main/src/colmap/exe/model.cc#L222-L262 ?

1. To the best of my knowledge, no. 2. The transformation represents a scaling, rotation, and translation.

It used to work on some earlier release. Unfortunately, I haven't had the time to figure out where the problem is and to fix it. Planning to do this on...

You are given an initial set of camera poses and intrinsics that might not be super accurate, so you want to refine them. This helps getting better poses and a...

Will have a look once I find some time on my hands

It seems that Colmap has an issue with the " " in the path to the file.

The two folders should contain the output of the reconstruction process, i.e., either the cameras.bin, images.bin, points3D.bin files or the cameras.txt, images.txt, points3D.txt files.