You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
You can find more visual results and a brief introduction to CASA at our project page.
Environmental Setup
The code is tested in:
Python 3.8
Pytorch 1.11.0
torchvision 0.12.0
Cuda 11.3
If you are using Anaconda, the following command can be used to build the environment:
conda env create -f casa.yml
conda activate casa
# install lietorch
git clone --recursive https://github.com/princeton-vl/lietorch.git
cd lietorch
python setup.py install
cd -
# install clip
pip install git+https://github.com/openai/CLIP.git
# install softras
# to compile for different GPU arch, see https://discuss.pytorch.org/t/compiling-pytorch-on-devices-with-different-cuda-capability/106409
pip install -e softras
Overview
We provide instructions for shape optimization on synthetic data.
Preparing Dataset
We use the Planetzoo dataset. If you would like to download the Planetzoo data, please fill out this google form. Then send both the google form and a proof of game purchase of planetzoo to us at casa_planetzoo@googlegroups.com. We will send you the link to download the data.
@inproceedings{wu2022casa,
title={{CASA}: Category-agnostic Skeletal Animal Reconstruction},
author={Yuefan Wu* and Zeyuan Chen* and Shaowei Liu and Zhongzheng Ren and Shenlong Wang},
booktitle={NeurIPS},
year={2022}
}
License
The data is released under Planetzoo Terms of Use, and the code is release under a non-comercial creative commons license.
About
code for Category-agnostic Skeletal Animal Reconstruction