Method

MonoDSSMs-M [MonoDSSMs-M]
[Anonymous Submission]

Submitted on 26 Jun. 2024 07:52 by
[Anonymous Submission]

Running time:0.02 s
Environment:1 core @ 2.5 Ghz (Python + C/C++)

Method Description:
An efficient monocular 3d object detection base on
Mamba with precise depth map as guidance for
learning process.
Parameters:
The model use 23M parameters.
Latex Bibtex:

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) 93.96 % 88.31 % 76.15 %
Car (Orientation) 93.32 % 86.83 % 74.62 %
Car (3D Detection) 19.80 % 14.15 % 11.56 %
Car (Bird's Eye View) 28.29 % 19.59 % 16.34 %
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