Implement gale shapley algorithm for label-matching

[jo-mueller]

Description

The preiously implemented strategy for label-matching based on picking the label with the respective highest intersection over union did not perform so nicely for complicated arrangements of overlapping labels. I implemented the Gale-Shapley algorithm to deal with such situations and I think it performs reasonably fast and well.

The previous version also removed the entries in the similarity matrix referring to the background. For the implementation of above algorithm it was easier to force-match the background in both images with wach other (setting similarity_matrix[0,0]=1.0 instead of performing matching operations on similarity_matrix[1:, 1:]. This required minor changes in the previous implementation of max_similarity(...) matching.

Note: The current implementation is rather long in terms of code - if you have suggestions for better implementations, let me know, I'll make this a draft in the meantime.

@zoccoler, feel free to have a look at the new code as well :)

Type of change

• Bug-fix
• New feature
• Breaking change
• Documentation update

Tests

• I adapted existing tests, because
• I added new tests to cover the code change
• All tests pass with my change

Final checks

• My change is the minimal possible work for the desired feature/fix
• I updated the documentation where necessary to cover the change

[haesleinhuepf]

Hi Johannes @jo-mueller ,

wow, cool stuff you're adding here. I have mostly questions about documentation.

Thanks!

Robert

[jo-mueller]

@haesleinhuepf I think tests should now cover the new algorithm and the docstrings are in order for good documentation. Feel free to merge if you think it's fit.

[haesleinhuepf]

Hi @jo-mueller we have now conflicts with the documentation. Could you try to fix it and/or tell me how to do this?

[jo-mueller]

Hey @haesleinhuepf ,

in this case the conflicts came from rather minor differences in the docstrings of some functions. I think the easiest approach to address these conflicts is the following:

• Resolve conflicts in script files (or generally, non-documentation files)
• For all other files, taking either version should be fine.
• Rebuild docs - this should change the documentation files to a state that allows merging without conflicts.

I fixed it for this case.

[jo-mueller]

I fixed it for this case.

For future reference: I think this is also a good reason why it could make sense to keep the documentation on a separate branch (github offers to default to gh-pages) through a github action such as this or this. Thus, we would only make sure to resolve conflicts between scripts on our branches and the documentation would be automatically synchronized.

[haesleinhuepf]

Hi Johannes @jo-mueller ,

awesome! I just created issue #29 so that we keep track of the documentation issue. Otherwise, I would merge. Do you think it's ready?

[jo-mueller]

Yes :)

[haesleinhuepf]

Merged. Thanks for sharing this amazing algorithm Johannes @jo-mueller !