Technical University of Denmark
Browse

sorry, we can't preview this file

fdi16_dataset.zip (6.88 GB)

3Shape FDI 16 Meshes from Intraoral Scans

Download (6.88 GB)
Version 2 2024-02-19, 11:36
Version 1 2023-07-19, 12:01
dataset
posted on 2023-07-19, 12:01 authored by Johan Ziruo YeJohan Ziruo Ye, Thomas Ørkild, Peter Lempel Søndergaard, Søren HaubergSøren Hauberg

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. Training and validation data stems from clinics based in Europe, while the test sets is a mix of North American and South American scans. 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.

History

ORCID for corresponding depositor