Method

MonoSample (DID-M3D) [MonoSample (DID-M3D)]
https://github.com/EE615/MonoSample

Submitted on 6 Jan. 2024 14:22 by
qiao junchao (Beijing jiaotong University)

Running time:0.2 s
Environment:1 core @ 2.5 Ghz (C/C++)

Method Description:
We have designed a data augmentation method for
monocular 3D object detection tasks, aimed at
enhancing the model's positional estimation
capabilities and alleviating overfitting due to
uncertainty loss. We experimented with this method
in DID-M3D and demonstrated in the validation set
that it can significantly improve model performance
Parameters:
\alpha=0.2
Latex Bibtex:
@ARTICLE{10556700,
author={Qiao, Junchao and Liu, Biao and Yang,
Jiaqi and Wang, Baohua and Xiu, Sanmu and Du, Xin
and Nie, XiaoBo},
journal={IEEE Robotics and Automation Letters},
title={MonoSample: Synthetic 3D Data
Augmentation Method in Monocular 3D Object
Detection},
year={2024},
volume={9},
number={8},
pages={7326-7332},
keywords={Three-dimensional
displays;Training;Object detection;Data
augmentation;Solid modeling;Uncertainty;Laser
radar;Computer vision for transportation;deep
learning for visual perception;object detection},
doi={10.1109/LRA.2024.3414272}}

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) 96.45 % 95.02 % 85.58 %
Car (Orientation) 96.30 % 94.69 % 85.10 %
Car (3D Detection) 28.63 % 18.05 % 15.19 %
Car (Bird's Eye View) 37.64 % 23.94 % 20.46 %
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