The KITTI Vision Benchmark Suite

Method

ARPNET(LiDAR-only) [ARPNET]

Submitted on 28 May. 2019 04:28 by
Yangyang Ye (Zhejiang University)

Running time:		0.08 s
Environment:		GPU @ 2.5 Ghz (Python + C/C++)

Method Description:

TBD

Parameters:

TBD

Latex Bibtex:

@Article{Ye2019,
author="Ye, Yangyang
and Zhang, Chi
and Hao, Xiaoli",
title="ARPNET: attention region proposal network
for 3D object detection",
journal="Science China Information Sciences",
year="2019",
month="Nov",
day="05",
volume="62",
number="12",
pages="220104",
issn="1869-1919",
doi="10.1007/s11432-019-2636-x",
url="https://doi.org/10.1007/s11432-019-2636-x"
}

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)	94.00 %	90.99 %	83.49 %
Car (Orientation)	93.42 %	90.11 %	82.56 %
Car (3D Detection)	84.69 %	74.04 %	68.64 %
Car (Bird's Eye View)	90.06 %	86.81 %	79.41 %
Pedestrian (Detection)	69.08 %	56.42 %	52.69 %
Pedestrian (Orientation)	60.47 %	48.49 %	45.02 %
Pedestrian (3D Detection)	48.32 %	39.31 %	35.93 %
Pedestrian (Bird's Eye View)	55.48 %	45.92 %	42.54 %
Cyclist (Detection)	84.96 %	71.95 %	65.21 %
Cyclist (Orientation)	82.61 %	68.72 %	62.00 %
Cyclist (3D Detection)	74.21 %	58.20 %	52.13 %
Cyclist (Bird's Eye View)	82.32 %	66.39 %	58.80 %