Method

Towards High Quality 3D Object Detection with Self-Attention [SA-Net]


Submitted on 11 Jan. 2023 07:21 by
xuchang li (CQUT)

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

Method Description:
Towards High Quality 3D Object Detection with Self-Attention
Parameters:
\alpha=0.2
Latex Bibtex:
\inproceedings{...}

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) 98.29 % 95.52 % 92.69 %
Car (Orientation) 98.25 % 95.33 % 92.39 %
Car (3D Detection) 91.37 % 84.39 % 77.61 %
Car (Bird's Eye View) 95.11 % 91.37 % 86.46 %
This table as LaTeX

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



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



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



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




eXTReMe Tracker