TAPoseNet

[MICCAI 2024] TAPoseNet: Teeth Alignment based on Pose estimation via multi-scale Graph Convolutional Network

Getting Started

You can run the given data in "dataset_test" directory. If you test your own data, please make sure that you have segmented the oral scan and ensure that the naming format of each tooth file matches the test data and that the overall coordinate system orientation of the teeth aligns with the test data.

Prerequisites

The project is based on python3.7, torch==1.10.2, the packages below are also needed

trimesh, vtk, open3d

Running the tests

You need to estimate the pose of teeth first. The pose of teeth are save in .npy file

python generate_pose.py

With the estimated pose of teeth, you can predict the teeth alignment target

python test.py