Version update 2024-02-16: Changed test data to also be from european scans.
The FDI 16 tooth mesh dataset including original meshes and exact data used in the original VF-Net paper. All scans are fully anonymized and donated by 3Shape.
The data contains 4844/1465/1423 meshes and point clouds for train/val/test respectively for deep/machine learning. All data stems from clinics based in Europe. The scans were collected using TRIOS scanners, and primarily the TRIOS 3 model. The data serves as a new and highly relevant use case for computer algorithms for 3D objects.
Each scan contain:
"_mesh.ply" suffix file: A PLY file of the original segmented tooth mesh
"_point_cloud.ply": The vertices and facet midpoints of the mesh as a point cloud
"_curvature_edgedistance.dat": A binary file including first the curvature of each point in the "_point_cloud.ply" and subsequently the djikstra distance to nearest edge vertex.
Funding
This work was funded by the Innovation Fund Denmark (1044-00172B) and the Novo Nordisk Foun-
dation through the Center for Basic Machine Learning Research in Life Science (NNF20OC0062606).
It also received funding from the European Research Council (ERC) under the European Union’s
Horizon 2020 research, innovation programme (757360). SH was supported in part by a research
grant (42062) from VILLUM FONDEN.