Towards Semantic Understanding of Surrounding Vehicular Maneuvers: A Panoramic Vision-Based Framework for Real-World Highway Studies

Publikation: Forskning - peer reviewKonferenceartikel i proceeding

Abstrakt

This paper proposes the use of multiple low-cost visual sensors to obtain a surround view of the ego-vehicle for semantic understanding. A multi-perspective view will assist the analysis of naturalistic driving studies (NDS), by automating the task of data reduction of the observed sequences into events. A user-centric vision-based framework is presented using a vehicle detector and tracker in each separate perspective. Multi-perspective trajectories are estimated and analyzed to extract 14 different events, including potential dangerous behaviors such as overtakes and cut-ins. The system is tested on ten sequences of real-world data collected on U. S. highways. The results show the potential use of multiple low-cost visual sensors for semantic understanding around the ego-vehicle.
Luk

Detaljer

This paper proposes the use of multiple low-cost visual sensors to obtain a surround view of the ego-vehicle for semantic understanding. A multi-perspective view will assist the analysis of naturalistic driving studies (NDS), by automating the task of data reduction of the observed sequences into events. A user-centric vision-based framework is presented using a vehicle detector and tracker in each separate perspective. Multi-perspective trajectories are estimated and analyzed to extract 14 different events, including potential dangerous behaviors such as overtakes and cut-ins. The system is tested on ten sequences of real-world data collected on U. S. highways. The results show the potential use of multiple low-cost visual sensors for semantic understanding around the ego-vehicle.
OriginalsprogEngelsk
Titel2016 IEEE Computer Vision and Pattern Recognition Workshops : Automatic Traffic Surveillance
ForlagIEEE
Publikationsdato19 dec. 2016
Sider1584-1591
ISBN (Trykt)978-1-5090-1438-5
ISBN (Elektronisk)978-1-5090-1437-8
DOI
StatusUdgivet - 19 dec. 2016
PublikationsartForskning
Peer reviewJa
Begivenhed2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops: Automated Traffic Surveillance - Las Vegas, USA
Varighed: 1 jul. 2016 → …

Workshop

Workshop2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops: Automated Traffic Surveillance
LandUSA
ByLas Vegas
Periode01/07/2016 → …

Download-statistik

Ingen data tilgængelig
ID: 232669279