Method

PointRGCN: Graph Convolution Networks for 3D Vehicles Detection Refinement [PointRGCN]


Submitted on 27 Nov. 2019 12:50 by
Jesus Zarzar (KAUST)

Running time:0.26 s
Environment:GPU @ V100 (Python)

Method Description:
Intra- and inter-proposal GCN networks.
Parameters:
5-layer R-GCN and 3-layer C-GCN.
Latex Bibtex:
@article{Zarzar2019PointRGCNGC,
title={PointRGCN: Graph Convolution Networks for
3D Vehicles Detection Refinement},
author={Jesus Zarzar and Silvio Giancola and
Bernard Ghanem},
journal={ArXiv},
year={2019},
volume={abs/1911.12236}
}

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) 97.51 % 92.33 % 87.07 %
Car (Orientation) 97.48 % 92.15 % 86.83 %
Car (3D Detection) 85.97 % 75.73 % 70.60 %
Car (Bird's Eye View) 91.63 % 87.49 % 80.73 %
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