PaCo: Parametric Point Cloud Completion

---

PaCo implements parametric completion, a new point cloud completion paradigm that recovers parametric primitives rather than individual points, for polygonal surface reconstruction.

🛠️ Setup

Prerequisites

Before you begin, ensure that your system has the following prerequisites installed:

• Conda
• CUDA Toolkit
• gcc & g++

The code has been tested with Python 3.10, PyTorch 2.6.0 and CUDA 11.8.

Installation

1. Clone the repository and enter the project directory:

   git clone https://github.com/parametric-completion/paco && cd paco

2. Install dependencies: Create a conda environment with all required dependencies:

   . install.sh

🚀 Usage

🎯 Training

Start training using one of the two parallelization:

Distributed Data Parallel (DDP):

CUDA_VISIBLE_DEVICES=0,1 ./scripts/train_ddp.sh

Data Parallel (DP):

CUDA_VISIBLE_DEVICES=0,1 ./scripts/train_dp.sh

⚙️ Available configurations

# check available configurations for training
python train.py --cfg job

# check available configurations for evaluation
python test.py --cfg job

Alternatively, review the main configuration file: conf/config.yaml.

🚧 TODOs

• Pretrained weights
• Dataset and evaluation script
• Hugging Face space

🎓 Citation

If you use PaCo in a scientific work, please consider citing the paper:

[paper]  [arxiv]

@InProceedings{chen2025paco,
    title={Parametric Point Cloud Completion for Polygonal Surface Reconstruction}, 
    author={Zhaiyu Chen and Yuqing Wang and Liangliang Nan and Xiao Xiang Zhu},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    year={2025}
}

🙏 Acknowledgements

Part of our implementation is based on the PoinTr repository. We appreciate the authors for open-sourcing their great work.