A universal total curvature estimation method that works for both triangle meshes and point clouds. For details, see the SIGGRAPH 2023 paper by Crane Chen under the supervision of Misha Kazhdan.
This codebase was developed on MacOS 12.6, adapting from the template code of open3d, open3d-cmake-find-package, original research code can be found here.
- open3d
- STL
- eigen for matrix data structures
- openmp for parallelization
- cgal for delaunay triangulation
On MacOS, just issue (make sure that Open3D and CGAL are already installed on your machine, use Homebrew to install CGAL on MacOS):
git clone https://github.com/HeCraneChen/open3d-discrete-total-curvature.git --recursive
cd open3d-discrete-total-curvature
cd build
./Draw
Compile
Compile this project from scratch using the standard cmake routine (make sure that Open3D and CGAL are already installed on your machine, use Homebrew to install CGAL on MacOS):
cd open3d-discrete-total-curvature
rm -r build
mkdir build
cd build
cmake ..
make
@inproceedings{10.1145/3587421.3595439,
author = {Chen, Crane He},
title = {Estimating Discrete Total Curvature with Per Triangle Normal Variation},
year = {2023},
isbn = {9798400701436},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3587421.3595439},
doi = {10.1145/3587421.3595439},
booktitle = {ACM SIGGRAPH 2023 Talks},
articleno = {56},
numpages = {2},
location = {Los Angeles, CA, USA},
series = {SIGGRAPH '23}
}