Method

Instance-Aware Feature Aggregation [IAFA]


Submitted on 7 Jul. 2020 06:31 by
Dingfu ZHOU (BAIDU)

Running time:0.04 s
Environment:1 core @ 2.5 Ghz (C/C++)

Method Description:
Instance aware feature aggregation
Parameters:
monocular one-stage detector
Latex Bibtex:
@inproceedings{zhou2020iafa,
title={IAFA: Instance-Aware Feature Aggregation
for 3D Object Detection from a Single Image},
author={Zhou, Dingfu and Song, Xibin and Dai,
Yuchao and Yin, Junbo and Lu, Feixiang and Liao,
Miao and Fang, Jin and Zhang, Liangjun},
booktitle={Proceedings of the Asian Conference on
Computer Vision},
year={2020}
}

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) 93.08 % 89.46 % 79.83 %
Car (Orientation) 92.96 % 89.14 % 79.40 %
Car (3D Detection) 17.81 % 12.01 % 10.61 %
Car (Bird's Eye View) 25.88 % 17.88 % 15.35 %
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