Geometric Transformer for Fast and Robust Point Cloud Registration

PyTorch implementation of the paper:

Geometric Transformer for Fast and Robust Point Cloud Registration.

Zheng Qin, Hao Yu, Changjian Wang, Yulan Guo, Yuxing Peng, and Kai Xu.

Introduction

Installation

Please use the following command for installation.

poetry install

Installation as package

Use pip to install:

pip install git+https://github.com/Tsapiv/GeoTransformerCPU.git

or poetry (add following to your project's pyproject.toml ):

[tool.poetry.dependencies]
# ...
geotransformer = { git = "https://github.com/Tsapiv/GeoTransformerCPU.git" }
# ...
[[tool.poetry.source]]
name     = "pytorch-cpu"
priority = "explicit"
url      = "https://download.pytorch.org/whl/cpu"

Citation

@inproceedings{qin2022geometric,
    title={Geometric Transformer for Fast and Robust Point Cloud Registration},
    author={Zheng Qin and Hao Yu and Changjian Wang and Yulan Guo and Yuxing Peng and Kai Xu},
    booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month={June},
    year={2022},
    pages={11143-11152}
}

Acknowledgements

• D3Feat
• PREDATOR
• RPMNet
• CoFiNet
• huggingface-transformer
• SuperGlue