You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, that's great news! My framework is mostly just educational (and trying to keep it as not too many lines), so I don't expect it to outperform proper bundle, but definitely something I'll check out.
Hey,
Thanks for the great work! Just want to say hi and let you know that we have just made it possible for running COLMAP BA with pycolmap and pyceres (colmap/colmap#2509), which can potentially open the door for building the optimization with pytorch as well. Might be interesting to substitute https://github.com/colmap/colmap/blob/main/pycolmap/custom_bundle_adjustment.py#L193-L205 and the
PyBundleAdjuster
(https://github.com/colmap/colmap/blob/main/pycolmap/custom_bundle_adjustment.py#L13) with pytorch to more easily integrate learning-based losses if we could make the optimizer achieve similar performance as in ceres : )Best,
Shaohui
The text was updated successfully, but these errors were encountered: