Method

UDeerPEP [UDeerPEP]
TBD

Submitted on 15 Feb. 2024 12:19 by
dong zichao (cuhk)

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

Method Description:
We designed a novel point painting method. Panoptic
segmentation is used to provide dense semantic
label to each lidar points. A language model based
point encoder is used as an adapter.
Parameters:
lr = 1e-2
Latex Bibtex:
@misc{dong2023pep,
title={PeP: a Point enhanced Painting method
for unified point cloud tasks},
author={Zichao Dong and Hang Ji and Xufeng
Huang and Weikun Zhang and Xin Zhan and Junbo
Chen},
year={2023},
eprint={2310.07591},
archivePrefix={arXiv},
primaryClass={cs.CV}
}

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) 98.42 % 97.57 % 95.08 %
Car (Orientation) 98.40 % 97.39 % 94.80 %
Car (3D Detection) 91.77 % 86.72 % 82.57 %
Car (Bird's Eye View) 95.34 % 93.40 % 89.07 %
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