VPC-Net: Completion of 3D Vehicles from MLS Point Clouds

VPC-Net: Completion of 3D Vehicles from MLS Point Clouds ISPRS Journal (IF = 7.319)

Yan Xia, Yusheng Xu, Cheng Wang, Uwe Stilla

Technical University of Munich, Xiamen University

Introduction

VPC-Net is a neural network to synthesize complete, dense, and uniform point clouds for vehicles from MLS data. The arXiv version of VPC-Net can be found here.

Citation

@article{xia2021vpc,
  title={VPC-Net: Completion of 3D vehicles from MLS point clouds},
  author={Xia, Yan and Xu, Yusheng and Wang, Cheng and Stilla, Uwe},
  journal={ISPRS Journal of Photogrammetry and Remote Sensing},
  volume={174},
  pages={166--181},
  year={2021},
  publisher={Elsevier}
}

This code is built using Tensorflow 1.12 with CUDA 9.0 and tested on Ubuntu 16.04 with Python 3.5.

Complie TF Operators

---

Please follow PointNet++ to compile TF operators.

Training

---

1. Download shapenet_car directory on Google Drive.
2. Run python train.py

Testing

---

1. ShapenNet-car completion
   • run python test.py.
2. KITTI completion
   • Download KITTI data from PCN project.
   • run python test_kitti.py.

Acknowledgements

Our implementations based on PCN repository.