Enhanced Shape Matching

[jlevy44]

Some ideas here (hu moments should already be an option for this pipeline):
https://www.learnopencv.com/shape-matching-using-hu-moments-c-python/
https://docs.opencv.org/master/d5/d45/tutorial_py_contours_more_functions.html

Perhaps, contour / convex hull calculation and then wasserstein distance estimation:
http://scipy.github.io/devdocs/generated/scipy.stats.wasserstein_distance.html

Other:
https://github.com/jl626/torch_shape_measure
http://www.kernel-operations.io/geomloss/_auto_examples/optimal_transport/plot_optimal_transport_2D.html?highlight=wasserstein

[jlevy44]

https://docs.scipy.org/doc/scipy/reference/generated/scipy.spatial.procrustes.html
https://medium.com/@olga_kravchenko/generalized-procrustes-analysis-with-python-numpy-c571e8e8a421

[jlevy44]

https://github.com/jlevy44/PathFlow-MixMatch/blob/master/pathflow_mixmatch/cli.py#L61

[jlevy44]

https://github.com/magicleap/SuperGluePretrainedNetwork
https://github.com/yewzijian/RPMNet
https://github.com/rusty1s/deep-graph-matching-consensus