Method

Homography Loss for Monocular 3D Object Detection [HomoLoss(imvoxelnet)]
https://github.com/gujiaqivadin/HomographyLoss

Submitted on 17 Oct. 2021 08:33 by
Jiaqi Gu (Zhejiang University)

Running time:0.20 s
Environment:1 core @ 2.5 Ghz (Python)

Method Description:
Homogrpahy Loss
Parameters:
TBD
Latex Bibtex:
@InProceedings{Gu_2022_CVPR,
author={Gu, Jiaqi and Wu, Bojian and Fan, Lubin
and Huang,
Jianqiang and Cao, Shen and Xiang, Zhiyu and Hua,
Xian-Sheng},
title={Homogrpahy Loss for Monocular 3D Object
Detection},
booktitle = {Proceedings of the IEEE/CVF
Conference on Computer Vision and Pattern
Recognition (CVPR)},
month = {June},
year = {2022},
}

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) 92.81 % 82.54 % 72.80 %
Car (Orientation) 91.94 % 80.67 % 70.64 %
Car (3D Detection) 20.10 % 12.99 % 10.50 %
Car (Bird's Eye View) 29.18 % 19.25 % 16.21 %
This table as LaTeX


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



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



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



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




eXTReMe Tracker