Improve logging for errors in masking during feature extraction#3034
Improve logging for errors in masking during feature extraction#3034
Conversation
Co-authored-by: Johannes Schönberger <jsch@demuc.de>
Co-authored-by: Johannes Schönberger <jsch@demuc.de>
ahojnnes
left a comment
There was a problem hiding this comment.
Thanks for your contribution. LGTM.
Head branch was pushed to by a user without write access
|
@Ambrosiussen There seems to be a bug in the logic, see: https://github.com/colmap/colmap/actions/runs/12295764100/job/34313358272?pr=3034 |
I think it actually exposed a bug! If you look at the dataset, and the arguments used for automatic reconstruction. No mask is supposed to be used at all! Yet somehow it passes this check which was already there :)
|
|
image_reader.cc:56 needs the following I would guess: |
|
@ahojnnes Any idea why these tests dont run? EDIT: And right after I comment this they magically start running :D |
|
They need to be approved by a maintainer for first time contributors. |
|
Thank you for your contribution. |
…ap#3034) The Masking logging is quite unhelpful currently. There is no indication if the mask arguments used in the Feature Extractor are correct, or if a mask for any given image is missing. This PR helps visualizing this. --------- Co-authored-by: Johannes Schönberger <jsch@demuc.de> Co-authored-by: Paul-Edouard Sarlin <15985472+sarlinpe@users.noreply.github.com>
The Masking logging is quite unhelpful currently.
There is no indication if the mask arguments used in the Feature Extractor are correct, or if a mask for any given image is missing.
This PR helps visualizing this.