Method

Sparse-to-Dense 3D Object Detector for Point Cloud [STD]
https://github.com/tomztyang/3DSSD

Submitted on 16 Nov. 2019 09:03 by
yang zetong (sddu)

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

Method Description:
TBD
Parameters:
TBD
Latex Bibtex:
@article{std2019yang,
author = {Zetong Yang and
Yanan Sun and
Shu Liu and
Xiaoyong Shen and
Jiaya Jia},
title = {{STD:} Sparse-to-Dense 3D Object Detector for
Point Cloud},
journal = {ICCV},
year = {2019},
url = {http://arxiv.org/abs/1907.10471},
archivePrefix = {arXiv},
eprint = {1907.10471},

}

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) 96.14 % 93.22 % 90.53 %
Car (3D Detection) 87.95 % 79.71 % 75.09 %
Car (Bird's Eye View) 94.74 % 89.19 % 86.42 %
Pedestrian (Detection) 68.33 % 55.04 % 50.85 %
Pedestrian (3D Detection) 53.29 % 42.47 % 38.35 %
Pedestrian (Bird's Eye View) 60.02 % 48.72 % 44.55 %
Cyclist (Detection) 83.99 % 71.63 % 64.92 %
Cyclist (3D Detection) 78.69 % 61.59 % 55.30 %
Cyclist (Bird's Eye View) 81.36 % 67.23 % 59.35 %
This table as LaTeX


2D object detection 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



2D object detection 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



2D object detection 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