RFFS-Net

Code for "Beyond single receptive field: A receptive field fusion-and-stratification network for airborne laser scanning point cloud classification"[arXiv] in ISPRS Journal of Photogrammetry and Remote Sensing (ISPRS Journal Ph & RS).

Overview

You can train or test RFFS-Net on ISPRS Vaihingen 3D, LASDU, and DFC2019 (US3D) dataset.

Dependencies

• Python 3.6
• PyTorch 1.7
• cuda 11

Datasets

• ISPRS Vaihingen 3D
• LSADU
• DFC2019 (US3D)

Usage

• Stage1 Install pointnet2-ops

  Install pointnet2-ops.

  cd pointnet2_ops_lib
  python setup.py install

• Stage2 Train

  Train our RFFS-Net.

  cd tools
  python train.py

• Stage3 Test

  Test our RFFS-Net.

  cd tools
  python test.py

References

This repo is built based on PointConv_PyTorch. Thanks for their great work!

BibTeX

If you find our work and this repository useful. Please consider giving a star :star: and citation 📚.

@article{mao2022beyond,
  title={Beyond single receptive field: A receptive field fusion-and-stratification network for airborne laser scanning point cloud classification},
  author={Mao, Yongqiang and Chen, Kaiqiang and Diao, Wenhui and Sun, Xian and Lu, Xiaonan and Fu, Kun and Weinmann, Martin},
  journal={ISPRS Journal of Photogrammetry and Remote Sensing},
  volume={188},
  pages={45--61},
  year={2022},
  publisher={Elsevier}
}