Bayesian Reasoning Using 3D Relations for Lane Marker Detection

Bart Boesman, Lars Baunegaard With Jensen, Emre Baseski, Nicolas Pugeault, Norbert Krüger

Publikation: Bidrag til bog/antologi/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

Resumé

We introduce a lane marker detection algorithm that integrates 3D attributes as well as 3D relations between local edges and semi-global contours in a Bayesian framework. The algorithm is parameter free and does not make use of any heuristic assumptions. The reasoning is based on the complete conditional probabilities of the different cues which are estimated from a training set. The importance of the individual visual cues can be computed using a standard measure and the cues can then be combined in an optimal way. In addition we show that when doing 3D reasoning, the uncertainties connected to the reconstruction process need to be taken into account to make the reasoning process more stable. The results are shown on a publicly available data set.
OriginalsprogEngelsk
TitelIkke angivet
Antal sider8
ForlagInstitut für Computergraphik
Publikationsdato2009
StatusUdgivet - 2009
BegivenhedVision, Modeling, and Visualization Workshop 2009 - Braunschweig, Tyskland
Varighed: 16. nov. 200918. nov. 2009

Konference

KonferenceVision, Modeling, and Visualization Workshop 2009
LandTyskland
ByBraunschweig
Periode16/11/200918/11/2009

Fingeraftryk

Uncertainty

Citer dette

Boesman, B., Jensen, L. B. W., Baseski, E., Pugeault, N., & Krüger, N. (2009). Bayesian Reasoning Using 3D Relations for Lane Marker Detection. I Ikke angivet Institut für Computergraphik.
Boesman, Bart ; Jensen, Lars Baunegaard With ; Baseski, Emre ; Pugeault, Nicolas ; Krüger, Norbert. / Bayesian Reasoning Using 3D Relations for Lane Marker Detection. Ikke angivet. Institut für Computergraphik, 2009.
@inproceedings{da1e34f0b1a011de9743000ea68e967b,
title = "Bayesian Reasoning Using 3D Relations for Lane Marker Detection",
abstract = "We introduce a lane marker detection algorithm that integrates 3D attributes as well as 3D relations between local edges and semi-global contours in a Bayesian framework. The algorithm is parameter free and does not make use of any heuristic assumptions. The reasoning is based on the complete conditional probabilities of the different cues which are estimated from a training set. The importance of the individual visual cues can be computed using a standard measure and the cues can then be combined in an optimal way. In addition we show that when doing 3D reasoning, the uncertainties connected to the reconstruction process need to be taken into account to make the reasoning process more stable. The results are shown on a publicly available data set.",
author = "Bart Boesman and Jensen, {Lars Baunegaard With} and Emre Baseski and Nicolas Pugeault and Norbert Kr{\"u}ger",
year = "2009",
language = "English",
booktitle = "Ikke angivet",
publisher = "Institut f{\"u}r Computergraphik",

}

Boesman, B, Jensen, LBW, Baseski, E, Pugeault, N & Krüger, N 2009, Bayesian Reasoning Using 3D Relations for Lane Marker Detection. i Ikke angivet. Institut für Computergraphik, Vision, Modeling, and Visualization Workshop 2009, Braunschweig, Tyskland, 16/11/2009.

Bayesian Reasoning Using 3D Relations for Lane Marker Detection. / Boesman, Bart; Jensen, Lars Baunegaard With; Baseski, Emre; Pugeault, Nicolas; Krüger, Norbert.

Ikke angivet. Institut für Computergraphik, 2009.

Publikation: Bidrag til bog/antologi/rapport/konference-proceedingKonferencebidrag i proceedingsForskningpeer review

TY - GEN

T1 - Bayesian Reasoning Using 3D Relations for Lane Marker Detection

AU - Boesman, Bart

AU - Jensen, Lars Baunegaard With

AU - Baseski, Emre

AU - Pugeault, Nicolas

AU - Krüger, Norbert

PY - 2009

Y1 - 2009

N2 - We introduce a lane marker detection algorithm that integrates 3D attributes as well as 3D relations between local edges and semi-global contours in a Bayesian framework. The algorithm is parameter free and does not make use of any heuristic assumptions. The reasoning is based on the complete conditional probabilities of the different cues which are estimated from a training set. The importance of the individual visual cues can be computed using a standard measure and the cues can then be combined in an optimal way. In addition we show that when doing 3D reasoning, the uncertainties connected to the reconstruction process need to be taken into account to make the reasoning process more stable. The results are shown on a publicly available data set.

AB - We introduce a lane marker detection algorithm that integrates 3D attributes as well as 3D relations between local edges and semi-global contours in a Bayesian framework. The algorithm is parameter free and does not make use of any heuristic assumptions. The reasoning is based on the complete conditional probabilities of the different cues which are estimated from a training set. The importance of the individual visual cues can be computed using a standard measure and the cues can then be combined in an optimal way. In addition we show that when doing 3D reasoning, the uncertainties connected to the reconstruction process need to be taken into account to make the reasoning process more stable. The results are shown on a publicly available data set.

M3 - Article in proceedings

BT - Ikke angivet

PB - Institut für Computergraphik

ER -

Boesman B, Jensen LBW, Baseski E, Pugeault N, Krüger N. Bayesian Reasoning Using 3D Relations for Lane Marker Detection. I Ikke angivet. Institut für Computergraphik. 2009