Accurate 3D Object Detection from Point Cloud Data using Bird’s Eye View Representations

Nerea Aranjuelo, Nerea Aranjuelo, Guus Engels, David Montero, David Montero, Marcos Nieto, Ignacio Arganda-Carreras, Ignacio Arganda-Carreras, Ignacio Arganda-Carreras, Luis Unzueta, Oihana Otaegui

2021

Abstract

In this paper, we show that accurate 3D object detection is possible using deep neural networks and a Bird’s Eye View (BEV) representation of the LiDAR point clouds. Many recent approaches propose complex neural network architectures to process directly the point cloud data. The good results obtained by these methods have left behind the research of BEV-based approaches. However, BEV-based detectors can take advantage of the advances in the 2D object detection field and need to handle much less data, which is important in real-time automotive applications. We propose a two-stage object detection deep neural network, which takes BEV representations as input and validate it in the KITTI BEV benchmark, outperforming state-of-the-art methods. In addition, we show how additional information can be added to our model to improve the accuracy of the smallest and most challenging object classes. This information can come from the same point cloud or an additional sensor’s data, such as the camera.

Download


Paper Citation


in Harvard Style

Aranjuelo N., Engels G., Montero D., Nieto M., Arganda-Carreras I., Unzueta L. and Otaegui O. (2021). Accurate 3D Object Detection from Point Cloud Data using Bird’s Eye View Representations. In Proceedings of the 13th International Joint Conference on Computational Intelligence (IJCCI 2021) - Volume 1: NCTA; ISBN 978-989-758-534-0, SciTePress, pages 246-253. DOI: 10.5220/0010688400003063


in Bibtex Style

@conference{ncta21,
author={Nerea Aranjuelo and Guus Engels and David Montero and Marcos Nieto and Ignacio Arganda-Carreras and Luis Unzueta and Oihana Otaegui},
title={Accurate 3D Object Detection from Point Cloud Data using Bird’s Eye View Representations},
booktitle={Proceedings of the 13th International Joint Conference on Computational Intelligence (IJCCI 2021) - Volume 1: NCTA},
year={2021},
pages={246-253},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0010688400003063},
isbn={978-989-758-534-0},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 13th International Joint Conference on Computational Intelligence (IJCCI 2021) - Volume 1: NCTA
TI - Accurate 3D Object Detection from Point Cloud Data using Bird’s Eye View Representations
SN - 978-989-758-534-0
AU - Aranjuelo N.
AU - Engels G.
AU - Montero D.
AU - Nieto M.
AU - Arganda-Carreras I.
AU - Unzueta L.
AU - Otaegui O.
PY - 2021
SP - 246
EP - 253
DO - 10.5220/0010688400003063
PB - SciTePress