Method

Homogeneous Sparse Fuse Network [HS-fusion]


Submitted on 27 May. 2024 07:49 by
yingjuan tang (Beijing institute of Technology)

Running time:- s
Environment:1 core @ 2.5 Ghz (Python)

Method Description:
This paper proposes a novel multi-modal framework,
Homogeneous Sparse Fusion (HS-Fusion)
Parameters:
we confine them to a specific range: [0, 70.4] for
the X axis, [−40, 40] for the Y axis, and [−3, 1]
for the Z axis. The input voxel dimensions are
established at (0.05, 0.05, 0.1) in size.
Latex Bibtex:
Under review

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) 98.33 % 95.48 % 92.66 %
Car (Orientation) 98.30 % 95.36 % 92.45 %
Car (3D Detection) 89.12 % 83.42 % 78.60 %
Car (Bird's Eye View) 93.77 % 90.95 % 87.79 %
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