This is research code I wrote based on the work presented in my paper
Golla, T., Kneiphof, T., Kuhlmann, H., Weinmann, M. and Klein, R. (2020), Temporal Upsampling of Point Cloud Sequences by Optimal Transport for Plant Growth Visualization. Computer Graphics Forum, 39: 167-179. https://doi.org/10.1111/cgf.14009
This code accepts two point cloud files and two label files as input and computes non-injective correspondces for them, using optimal transport. Execute python ot-correspondences.py and supply the point cloud files and label files as input, as well as the output file name. The label files indicate segments of the point clouds per point. Correspondences are computed for point cloud segments with the same label.