mugdxy / 3d-voronoi-gpu

Source code for paper "Parallel Computation of 3D Clipped Voronoi Diagrams"

Parallel Computation of 3D Clipped Voronoi Diagrams

By Xiaohan Liu, Lei Ma, Jianwei Guo, Dong-Ming Yan

Build

Prerequisites:

• Windows 10 or Ubuntu 16.04+
• CUDA 10.1
• CMake

Please build our code with CMake.

For Linux, use the following steps:

mkdir build
cd build
cmake ..
make

Usage

./bin/VolumeVoronoiGPU <tet_mesh.tet/vtk> <sites_file.xyz> <nb_iter> <k (optional)>

Example:

./bin/VolumeVoronoiGPU ../data/joint.tet ../data/joint.xyz 1 0

It produces out.xyz file with barycenters per cell and record.csv file with performance data.

Arguments:

• tet_mesh: input tetrahedral mesh file (tet or vtk)
• sites_file: input sites file (xyz)
• nb_iter: the number of Lloyd's iteration
• k: use the given value if specified, else default value

Citation

If you find this work is useful for your research, please cite our paper:

@article{Liu2020Parallel,
  author={Liu, Xiaohan and Ma, Lei and Guo, Jianwei and Yan, Dong-Ming},
  journal={IEEE Transactions on Visualization and Computer Graphics}, 
  title={Parallel Computation of 3D Clipped Voronoi Diagrams}, 
  year={2020},
  volume={},
  number={},
  pages={1-1},
}

Acknowledgment

We thank Ray et al. for sharing their code.

Additional

This version is modified by MuGdxy

1. We added some tool scripts e.g. to normalize a tetrahedra mesh manually
2. We added some visualization tools (with Houdini)