BushNet: Effective semantic segmentation of bush in large-scale point clouds

Hejun Wei, Enyong Xu, Jinlai Zhang, Yanmei Meng, Jin Wei, Zhen Dong, Zhengqiang LiComput. Electron. Agric 2022 · DOI · OpenReview

Abstract

Highlights•A minimum probability random sampling method is designed.•A local multi-dimensional feature fusion module that can capture the point cloud features of bushes is proposed.•The multi-channel attention module was used to learn interactive information between channels.•The algorithm achieves the best balance of accuracy, speed and computational cost.
BibTeX
@article{wei2022bushnet,
  author = {Hejun Wei and Enyong Xu and Jinlai Zhang and Yanmei Meng and Jin Wei and Zhen Dong and Zhengqiang Li},
  title = {{BushNet: Effective semantic segmentation of bush in large-scale point clouds}},
  journal = {Comput. Electron. Agric.},
  year = {2022},
  doi = {10.1016/j.compag.2021.106653},
}

Loading related content...

Figure