Dataset is structured under 6 data parts. It is required to download all of them and merge them to a same folder architecture.

url : https://osf.io/xctdy/
Two dataset train/test splits are provided , which specify the samples to consider for each dataset:

3D Teeth Seg Challenge split (used during the challenge)
Teeth3DS official dataset split