Points2Poly is the implementation of the compact building surface reconstruction method described in this arxiv paper. This implementation incorporates learnable implicit surface representation into explicitly constructed geometry.
Due to clutter concerns, the core module is separately maintained in the abspy repository (also available as a PyPI package), while this repository acts as a wrapper with additional sources and instructions in particular for building reconstruction. The wrapper code is being cleaned.
The requirements consist of two parts: one from abspy
that provides functionalities on vertex group, cell complex, and adjacency graph; the other one from points2surf
that facilitates occupancy estimation.
Follow this instruction to install abspy
with its appendencies.
Init points2surf
submodule:
git submodule init
Install the requirements listed in points2surf/requirements.txt
with PyPI:
pip install -r points2surf/requirements.txt
For training the neural network, make sure CUDA is available and enabled. You can navigate to the points2surf repository for more details on its requirements.
- Separate
abspy
/points2surf
frompoints2poly
wrappers - Config with hydra
- Tutorial on how to get started
- Host generated data
If you use abspy in a scientific work, please cite:
@article{chen2021reconstructing,
title={Reconstructing Compact Building Models from Point Clouds Using Deep Implicit Fields},
author={Chen, Zhaiyu and Khademi, Seyran and Ledoux, Hugo and Nan, Liangliang},
journal={arXiv preprint arXiv:2112.13142},
year={2021}
}