/RandLA-Net

RandLA-Net in Tensorflow

MIT LicenseMIT

PWC

RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds

Qingyong Hu, Bo Yang, Linhai Xie, Stefano Rosa, Yulan Guo, Zhihua Wang, Niki Trigoni, Andrew Markham. arXiv:1911.11236, 2019. RandLA-Net in Tensorflow, coming soon

Introduction

This repository contains the implementation of RandLA-Net, a simple and efficient neural architecture for semantic segmentation of large-scale 3D point clouds.

The following figure shows the basic building block of our RandLA-Net:

Qualitative Results

S3DIS

2 z

Semantic3D

z z

SemanticKITTI

zzz

Quantitative Results

Semantic3D

Quantitative results of different approaches on Semantic3D (reduced-8). Only the recent published approaches are compared. Accessed on 15 November 2019.

a

SemanticKITTI

Quantitative results of different approaches on SemanticKITTI dataset.

s

S3DIS

Quantitative results of different approaches on S3DIS dataset.

Demo

Citation

If you find our work useful in your research, please consider citing:

@article{hu2019randla,
  title={RandLA-Net: Efficient Semantic Segmentation of Large-Scale Point Clouds},
  author={Hu, Qingyong and Yang, Bo and Xie, Linhai and Rosa, Stefano and Guo, Yulan and Wang, Zhihua and Trigoni, Niki and Markham, Andrew},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  year={2020}
}