This is the code of the framework for multimodal 3D registration of CT volumes and 3D point clouds which has been proposed in 'Multimodal registration across 3D point clouds and CT-volumes' (Multimodal registration across 3D point clouds and CT-volumes - ScienceDirect): Saiti, E., and T. Theoharis. "Multimodal registration across 3D point clouds and CT-volumes." Computers & Graphics 106 (2022): 259-266. It is A deep learning network that combines regular CNNs suited for data with standard grid structure and geometric deep learning suited for unstructured data.