Shining3D tooth pose dataset

The training set, validation set, and testing set in the constructed Shining3D tooth pose dataset contain 1689, 150, and 150 samples, respectively. Jaw models are generated from hospital patients by oral scanning. The ground truth of the relative pose of each tooth is generated by adding random jittering to the tooth models. For each tooth, ground truth relative pose information was generated by introducing random jittering to the tooth models. For jaw and tooth pose estimation, we utilize the ADD-S and ADD(-S) metrics, as well as their corresponding area under the curve (AUC), as the evaluation criteria. This dataset can be used for  multiview object pose estimation.