Occlusion aware sensor fusion for early crossing pedestrian detection

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

12 Citations (Scopus)
255 Downloads (Pure)

Abstract

Early and accurate detection of crossing pedestrians is crucial in automated driving to execute emergency manoeuvres in time. This is a challenging task in urban scenarios however, where people are often occluded (not visible) behind objects, e.g. other parked vehicles. In this paper, an occlusion aware multi-modal sensor fusion system is proposed to address scenarios with crossing pedestrians behind parked vehicles. Our proposed method adjusts the detection rate in different areas based on sensor visibility. We argue that using this occlusion information can help to evaluate the measurements. Our experiments on real world data show that fusing radar and stereo camera for such tasks is beneficial, and that including occlusion into the model helps to detect pedestrians earlier and more accurately.
Original languageEnglish
Title of host publicationProceedings IEEE Symposium Intelligent Vehicles (IV 2019, Paris)
Place of PublicationPiscataway, NJ, USA
PublisherIEEE
Pages1768-1774
ISBN (Electronic)978-1-7281-0560-4
DOIs
Publication statusPublished - 2019
EventIEEE Intelligent Vehicles Symposium 2019 - Paris, France
Duration: 9 Jun 201912 Jun 2019

Conference

ConferenceIEEE Intelligent Vehicles Symposium 2019
Abbreviated titleIV 2019
Country/TerritoryFrance
CityParis
Period9/06/1912/06/19

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Fingerprint

Dive into the research topics of 'Occlusion aware sensor fusion for early crossing pedestrian detection'. Together they form a unique fingerprint.

Cite this