Shining3D tooth segmentation dataset

The Shining3D dataset consists of 1866, 272, and 272 videos for the training, validation, and testing sets, respectively. The uniform size of the images in the dataset is 640 × 480 pixels.  LabelMe software is employed to accurately mark the boundary and classify the region of each tooth in the datasets.  This dataset is usef for orthodontic treatment. Orthodontic treatment monitoring, one of the research direction of artificial intelligence, involves using current images and previous 3D models to estimate the relative position of individual teeth before and after orthodontic treatment. This dataset can be used to varify the effectiveness and efficiency of the proposed methods.