8000 GitHub - bibi547/TSRNet: TSRNet: A Dual-stream Network for Refining 3D Tooth Segmentation [TVCG 2024]
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

bibi547/TSRNet

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

20 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

TSRNet: A Dual-stream Network for Refining 3D Tooth Segmentation

Run

1. Mesh Simplification

Simplify original tooth meshes to about 10,000 facets. Mesh simplification can be achieved using functions from Open3D or through implementations by MeshCNN, which utilize the bpy library.

2. Coarse Segmentation Methods

DGCNN, TeethGNN, or others.

3. Extracting Geodesic Distance Maps

./scripts/geodesic_distance.py

Geodesic distance maps are extracted during the data preprocessing phase. Based on the ground truth and the coarse segmentation results, extract the segmentation boundaries (vertices). Then, compute the shortest geodesic distance from mesh vertices to the segmentation boundaries and save the results as '.txt' files.

4. Config

Modify the config and dataset based on the path and filename.

5. Requirement

...

6. Train

python train.py

7. Test

python test.py

Citation

@article{jin2024tsrnet,
  title={TSRNet: A Dual-stream Network for Refining 3D Tooth Segmentation},
  author={Jin, Hairong and Shen, Yuefan and Lou, Jianwen and Zhou, Kun and Zheng, Youyi},
  journal={IEEE Transactions on Visualization and Computer Graphics},
  year={2024},
  publisher={IEEE}
}

Acknowledgement

MeshCNN

DGCNN

Point Transformer

Teeth3DS dataset

About

TSRNet: A Dual-stream Network for Refining 3D Tooth Segmentation [TVCG 2024]

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

0