Instructions and scripts for rendering point cloud figures shown in PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flows.
The script generates a XML file, which describes a 3D scene in the format used by Mitsuba. You will then be able to render it with Mitsuba.
- Python 3.6+
- Mitsuba Renderer 2
bash install_mitsuba.sh
# add mitsuba to path
export PATH=$PWD/mitsuba2/build/dist/:$PATH
# Generate scene XML file
DATA_PATH=chair_pcl.npy
SAVE_PATH=/D_data/tiankai/pc_viz
python pointflow_fig_colorful.py $DATA_PATH $SAVE_PATH
Please consider citing our work if you find it useful:
@article{pointflow,
title={PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flows},
author={Yang, Guandao and Huang, Xun, and Hao, Zekun and Liu, Ming-Yu and Belongie, Serge and Hariharan, Bharath},
journal={arXiv},
year={2019}
}