KITTI-360

Method

PointNet++ [PointNet++]
https://github.com/yanx27/Pointnet_Pointnet2_pytorch

Submitted on 22 Oct. 2021 13:01 by
Yiyi Liao (MPI)

Running time:
Environment:NVIDIA V100

Method Description:
A baseline method for 3D semantic segmentation. Trained and evaluated on KITTI-360.
Parameters:
Latex Bibtex:
@InProceedings{Qi2017NeurIPS,
author = {Qi, Charles R and Yi, Li and Su, Hao and Guibas, Leonidas J},
title = {PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space},
booktitle = NeurIPS,
year = {2017},
}

Test Set Average

road sidewalk building wall fence pole traffic light traffic sign vegetation terrain person car truck motorcycle bicycle mIoU class
82.06 66.29 62.11 30.48 24.92 38.26 0.00 23.38 71.19 47.29 2.05 84.77 0.49 1.64 0.00 35.66
flat construction object nature human vehicle mIoU category
87.23 59.58 40.56 76.33 2.05 83.91 58.28
This table as LaTeX




eXTReMe Tracker