Publication:
BirdNet+: two-stage 3D object detection in LiDAR through a sparsity-invariant bird's eye view

Simple item page
dc.affiliation.dptoUC3M. Departamento de Ingeniería de Sistemas y Automáticaes
dc.affiliation.dptoUC3M. Departamento de Informáticaes
dc.affiliation.grupoinvUC3M. Grupo de Investigación: Laboratorio de Sistemas Inteligenteses
dc.affiliation.grupoinvUC3M. Grupo de Investigación: Laboratorio de Control, Aprendizaje y Optimización de Sistemas (CAOS)es
dc.contributor.authorBarrera Del Pozo, Alejandro
dc.contributor.authorBeltrán de la Cita, Jorge
dc.contributor.authorGuindel Gómez, Carlos
dc.contributor.authorIglesias Martínez, José Antonio
dc.contributor.authorGarcía Fernández, Fernando
dc.contributor.funderComunidad de Madrides
dc.date.accessioned2023-02-03T13:51:10Z
dc.date.available2023-02-03T13:51:10Z
dc.date.issued2021-11-30
dc.description.abstractAutonomous navigation relies upon an accurate understanding of the elements in the surroundings. Among the different on-board perception tasks, 3D object detection allows the identification of dynamic objects that cannot be registered by maps, being key for safe navigation. Thus, it often requires the use of LiDAR data, which is able to faithfully represent the scene geometry. However, although raw laser point clouds contain rich features to perform object detection, more compact representations such as the bird's eye view (BEV) projection are usually preferred in order to meet the time requirements of the control loop. This paper presents an end-to-end object detection network based on the well-known Faster R-CNN architecture that uses BEV images as input to produce the final 3D boxes. Our regression branches can infer not only the axis-aligned bounding boxes but also the rotation angle, height, and elevation of the objects in the scene. The proposed network provides state-of-the-art results for car, pedestrian, and cyclist detection with a single forward pass when evaluated on the KITTI 3D Object Detection Benchmark, with an accuracy that exceeds 64% mAP 3D for the Moderate difficulty. Further experiments on the challenging nuScenes dataset show the generalizability of both the method and the proposed BEV representation against different LiDAR devices and across a wider set of object categories by being able to reach more than 30% mAP with a single LiDAR sweep and almost 40% mAP with the usual 10-sweep accumulation.en
dc.description.sponsorshipThis work was supported in part by the Government of Madrid (Comunidad de Madrid) under the Multiannual Agreement with the University Carlos III of Madrid (UC3M) in the line of "Fostering Young Doctors Research"(PEAVAUTO-CM-UC3M), and in part by the Context of the V Regional Programme of Research and Technological Innovation (PRICIT).en
dc.format.extent18
dc.identifier.bibliographicCitationBarrera, A., Beltran, J., Guindel, C., Iglesias, J. A. & Garcia, F. (2021). BirdNet+: Two-Stage 3D Object Detection in LiDAR Through a Sparsity-Invariant Bird’s Eye View. IEEE Access, 9, 160299-160316.en
dc.identifier.doihttps://doi.org/10.1109/ACCESS.2021.3131389
dc.identifier.publicationfirstpage160299
dc.identifier.publicationlastpage160316
dc.identifier.publicationtitleIEEE Accessen
dc.identifier.publicationvolume9
dc.identifier.urihttps://hdl.handle.net/10016/36461
dc.identifier.uxxiAR/0000030633
dc.language.isoeng
dc.publisherIEEE
dc.relation.projectIDComunidad de Madrid. PEAVAUTO-CM-UC3Mes
dc.rights© The authors, 2021en
dc.rightsAtribución 3.0 España*
dc.rights.accessRightsopen accessen
dc.rights.urihttp://creativecommons.org/licenses/by/3.0/es/*
dc.subject.ecienciaInformáticaes
dc.subject.ecienciaIngeniería Mecánicaes
dc.subject.ecienciaRobótica e Informática Industriales
dc.subject.otherBird's eye view (BEV)en
dc.subject.otherLiDARen
dc.subject.otherObject detectionen
dc.subject.otherAutonomous drivingen
dc.titleBirdNet+: two-stage 3D object detection in LiDAR through a sparsity-invariant bird's eye viewen
dc.typeresearch article*
dc.type.hasVersionVoR*
dspace.entity.typePublication
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
BirdNet(plus)_IEEEA_2021.pdf
Size:
3.77 MB
Format:
Adobe Portable Document Format
Download
Collections
DISA - LSI - Artículos de Revistas
DI - CAOS - Artículos de Revistas