Deep Reconstruction of 3D Smoke Densities from Artist Sketches

PyTorch implementation of Deep Reconstruction of 3D Smoke Densities from Artist Sketches.

Byungsoo Kim, Xingchang Huang, Laura Wuelfroth, Jingwei Tang, Guillaume Cordonnier, Markus Gross, Barbara Solenthaler

ETH Zurich, Max Planck Institute for Informatics, Inria d'Université Côte d'Azur

Computer Graphics Forum (Proceedings of Eurographics 2022)

Install

We strongly recommend to use a conda enviroment to develop this package. If you don't have one create it.

  # install conda
  wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh
  chmod +x Miniconda-latest-Linux-x86_64.sh
  ./Miniconda3-latest-Linux-x86_64.sh

  # update conda
  conda update -y conda
  
  # create conda environment
  conda env create -f project_env.yml
  conda activate sketch2fluid

Houdini Plugin

  git checkout houdini_plugin

Original Repository

https://gitlab.ethz.ch/cglsim/sketch2fluid