Decimate returning Fine to Coarse Map

Repository wrapping decimate function of libigl to additionally compute a fine to coarse map for triangles and vertices. With your full resolution mesh be V, F and your decimated mesh be Vd, Fd then you obtain maps I: V -> Vd and J: F -> Fd (contrary to decimate function in libigl which returns maps I: Vd -> V and J: Fd -> F).

Note: this algorithm might be very slow for large shapes since the igl:decimate function is called 250 times assuming you reduce the shape by 500 triangles.

Note: this algorithm uses the old libigl decimate functions (which do NOT avoid intersections of the shapes) and thus result in nicer triangulation.

⚙️ Installation

This repo comes with c++ code, python code and a matlab mex wrapper 💡

Note: Code only tested on unix systems. Expect issues with Windows 🪟.

🐍 Python

Simply download the repo and python setup.py install (you need working c++ compiler + cmake installed on your machine).

🔺 Matlab

See example file BUILD_MEX.sh on how to compile matlab mex wrapper. ⚠️ Do not simply run the .sh file as you need to adjust the variables Matlab_MEX_EXTENSION and Matlab_ROOT_DIR depending on your system.

🚀 Cpp

Simply include the singlemost header decimate.hpp file of this project into your project.

✨ Usage

🐍 Python

import fine2coarsedec as f2cd
target_num_faces = 100
use_qslim = False # choose different decimation algorithm
vred, fred, ff2c, vf2c = f2cd.decimate(v, f, target_num_faces, use_qslim)

# in case you only need decimated shapes without fine to coarse map (which is faster)
vred, fred = f2cd.decimate_no_map(v, f, target_num_faces, use_qslim)

🔺 Matlab

target_num_faces = 100
use_qslim = false
[vred, fred, ff2c, vf2c] = f2cdecimate(v, f, target_num_faces, use_qslim);

🚀 Cpp

#include "include/decimate.hpp"
Eigen::VectorXi Vfine2coarse, Ffine2coarse;
const bool useQslim = true;
const int targetNumFaces = 100;
f2cdec::decimator(V, F, targetNumFaces, Vfine2coarse, Ffine2coarse, useqslim);

Attribution 🎓

When using this code for your own projects please cite the followig:

@misc{decimatec2f,
  title = {Decimate Coarse to Fine},
  author = {Paul Roetzer and others},
  note = {https://github.com/paul0noah/c2fdecimate},
  year = {2023},
}

License 🚀

This repo is licensed under MIT licence.