About the function pycolmap.triangulate_points
When I run a command
pycolmap.triangulate_points( reference_model, database_path, image_dir, model_path, options=options )
Are the map points triangulated only? Will the known camera intrinsic parameters and camera poses be optimized?
Looking forward to your answer!
I think so, as the source C++ code for pycolmap.triangulate_points seems to be this:
https://github.com/colmap/colmap/blob/dcfd14b300868c7eb1b360ebbca0dc21acd641b6/src/colmap/controllers/incremental_mapper.cc#L540
Camera poses are always kept fixed, intrinsics are optimized only if refine_intrinsics=True.
https://github.com/colmap/colmap/blob/03556bb2cc941b58c3a4c244e5549e06c2f8117a/src/colmap/exe/sfm.cc#L446-L449