"If your pair of jeans isn't even isotropic, why should its parameterization be?" Previous UV mapping methods (ABF++, LSCM, ARAP, BFF, SCAF, AutoCuts, OptCuts, ...) are rotation invariant and minimize triangle distortion without any assumption on the material being flattened. These methods are unfit for anisotropic materials which stretch unequally along different axes. This repository provides an implementation of the anisotropic parameterization described in Computational Pattern Making from 3D Garment Models by Pietroni et al. It's original intent is to accurately flatten woven textiles, for which the thread structure induces anisotropy in their ability to stretch. | ![]() |
git clone https://github.com/CorentinDumery/garment-flattening
cd garment-flattening/
git submodule update --init --recursive
mkdir build && cd build
cmake ..
make -j woven_param #build library
make -j woven_viz #build test app
./woven_viz [path/to/input/mesh.obj]
Then, the main features of our parameterization can be visualized with the following commands:
make -j woven_viz && ./woven_viz ../data/jumpsuit_multipose/front1.obj
make -j multiple_poses && ./multiple_poses
make -j reflec_param && ./reflec_param
git add submodule https://github.com/CorentinDumery/garment-flattening
git submodule update --init --recursive
Then, in your CMakeLists.txt
, add:
add_subdirectory([path_to_garment-flattening]/garment-flattening)
...
target_link_libraries([target] woven_param)
Thank you for reading! If this repository is useful to you, feel free to reach out and/or cite our paper:
@article{ComputationalPatternmaking:2022,
title = {Computational Pattern Making from {3D} Garment Models},
author = {Nico Pietroni and Corentin Dumery and Raphael Falque and Mark Liu and Teresa Vidal-Calleja and Olga Sorkine-Hornung},
journal = {ACM Transactions on Graphics},
volume = {41},
number = {4},
pages = {157:1–14},
year = {2022},
publisher = {ACM}
}
Some of the models used in this repository are adapted from Generating Datasets of 3D Garments with Sewing Patterns by Maria Korosteleva and Sung-Hee Lee.