Method

3D IoU-Net [3D IoU-Net]


Submitted on 13 Feb. 2020 06:05 by
Jiale Li (Zhejiang University)

Running time:0.1 s
Environment:1 core @ 2.5 Ghz (C/C++)

Method Description:
To be updated.
Parameters:
alpha=0.2
Latex Bibtex:
@article{Li20203DIoUNet,
title={3D IoU-Net: IoU Guided 3D Object Detector for
Point Clouds},
author={Li, Jiale and Luo, Shujie and Zhu, Ziqi and
Dai, Hang and Krylov, S. Andrey and Ding, Yong and
Shao, Ling},
journal={arXiv preprint arXiv:2004.04962},
year={2020},
}

Detailed Results

Object detection and orientation estimation results. Results for object detection are given in terms of average precision (AP) and results for joint object detection and orientation estimation are provided in terms of average orientation similarity (AOS).


Benchmark Easy Moderate Hard
Car (Detection) 96.31 % 92.47 % 87.67 %
Car (Orientation) 96.31 % 92.42 % 87.60 %
Car (3D Detection) 87.96 % 79.03 % 72.78 %
Car (Bird's Eye View) 94.76 % 88.38 % 81.93 %
This table as LaTeX


2D object detection results.
This figure as: png eps pdf txt gnuplot



Orientation estimation results.
This figure as: png eps pdf txt gnuplot



3D object detection results.
This figure as: png eps pdf txt gnuplot



Bird's eye view results.
This figure as: png eps pdf txt gnuplot




eXTReMe Tracker