Scene Flow Estimation from Light Fields via the Preconditioned Primal-Dual Algorithm

Stefan Heber, Thomas Pock

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In this paper we present a novel variational model to jointly estimate geometry and motion from a sequence of light fields captured with a plenoptic camera. The proposed model uses the so-called sub-aperture representation of the light field. Sub-aperture images represent images with slightly different viewpoints, which can be extracted from the light field. The sub-aperture representation allows us to formulate a convex global energy functional, which enforces multi-view geometry consistency, and piecewise smoothness assumptions on the scene flow variables. We optimize the proposed scene flow model by using an efficient preconditioned primal-dual algorithm. Finally, we also present synthetic and real world experiments.
LanguageEnglish
Title of host publicationPattern Recognition: 36th German Conference, GCPR 2014, M{\"u}nster, Germany, September 2-5, 2014, Proceedings
PublisherSpringer International Publishing AG
Pages3-14
Volume8753
ISBN (Electronic)978-3-319-11752-2
ISBN (Print)978-3-319-11751-5
DOIs
StatusAccepted/In press - 2014

Fingerprint

Geometry
Cameras
Experiments

Fields of Expertise

  • Information, Communication & Computing

Cite this

Heber, S., & Pock, T. (2014). Scene Flow Estimation from Light Fields via the Preconditioned Primal-Dual Algorithm. In Pattern Recognition: 36th German Conference, GCPR 2014, M{\"u}nster, Germany, September 2-5, 2014, Proceedings (Vol. 8753, pp. 3-14). Springer International Publishing AG . DOI: 10.1007/978-3-319-11752-2_1

Scene Flow Estimation from Light Fields via the Preconditioned Primal-Dual Algorithm. / Heber, Stefan; Pock, Thomas.

Pattern Recognition: 36th German Conference, GCPR 2014, M{\"u}nster, Germany, September 2-5, 2014, Proceedings. Vol. 8753 Springer International Publishing AG , 2014. p. 3-14.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Heber, S & Pock, T 2014, Scene Flow Estimation from Light Fields via the Preconditioned Primal-Dual Algorithm. in Pattern Recognition: 36th German Conference, GCPR 2014, M{\"u}nster, Germany, September 2-5, 2014, Proceedings. vol. 8753, Springer International Publishing AG , pp. 3-14. DOI: 10.1007/978-3-319-11752-2_1
Heber S, Pock T. Scene Flow Estimation from Light Fields via the Preconditioned Primal-Dual Algorithm. In Pattern Recognition: 36th German Conference, GCPR 2014, M{\"u}nster, Germany, September 2-5, 2014, Proceedings. Vol. 8753. Springer International Publishing AG . 2014. p. 3-14. Available from, DOI: 10.1007/978-3-319-11752-2_1
Heber, Stefan ; Pock, Thomas. / Scene Flow Estimation from Light Fields via the Preconditioned Primal-Dual Algorithm. Pattern Recognition: 36th German Conference, GCPR 2014, M{\"u}nster, Germany, September 2-5, 2014, Proceedings. Vol. 8753 Springer International Publishing AG , 2014. pp. 3-14
@inproceedings{cac52360ee914b518ec058d2919aede1,
title = "Scene Flow Estimation from Light Fields via the Preconditioned Primal-Dual Algorithm",
abstract = "In this paper we present a novel variational model to jointly estimate geometry and motion from a sequence of light fields captured with a plenoptic camera. The proposed model uses the so-called sub-aperture representation of the light field. Sub-aperture images represent images with slightly different viewpoints, which can be extracted from the light field. The sub-aperture representation allows us to formulate a convex global energy functional, which enforces multi-view geometry consistency, and piecewise smoothness assumptions on the scene flow variables. We optimize the proposed scene flow model by using an efficient preconditioned primal-dual algorithm. Finally, we also present synthetic and real world experiments.",
author = "Stefan Heber and Thomas Pock",
year = "2014",
doi = "10.1007/978-3-319-11752-2_1",
language = "English",
isbn = "978-3-319-11751-5",
volume = "8753",
pages = "3--14",
booktitle = "Pattern Recognition: 36th German Conference, GCPR 2014, M{\{"}u}nster, Germany, September 2-5, 2014, Proceedings",
publisher = "Springer International Publishing AG",
address = "Switzerland",

}

TY - GEN

T1 - Scene Flow Estimation from Light Fields via the Preconditioned Primal-Dual Algorithm

AU - Heber,Stefan

AU - Pock,Thomas

PY - 2014

Y1 - 2014

N2 - In this paper we present a novel variational model to jointly estimate geometry and motion from a sequence of light fields captured with a plenoptic camera. The proposed model uses the so-called sub-aperture representation of the light field. Sub-aperture images represent images with slightly different viewpoints, which can be extracted from the light field. The sub-aperture representation allows us to formulate a convex global energy functional, which enforces multi-view geometry consistency, and piecewise smoothness assumptions on the scene flow variables. We optimize the proposed scene flow model by using an efficient preconditioned primal-dual algorithm. Finally, we also present synthetic and real world experiments.

AB - In this paper we present a novel variational model to jointly estimate geometry and motion from a sequence of light fields captured with a plenoptic camera. The proposed model uses the so-called sub-aperture representation of the light field. Sub-aperture images represent images with slightly different viewpoints, which can be extracted from the light field. The sub-aperture representation allows us to formulate a convex global energy functional, which enforces multi-view geometry consistency, and piecewise smoothness assumptions on the scene flow variables. We optimize the proposed scene flow model by using an efficient preconditioned primal-dual algorithm. Finally, we also present synthetic and real world experiments.

U2 - 10.1007/978-3-319-11752-2_1

DO - 10.1007/978-3-319-11752-2_1

M3 - Conference contribution

SN - 978-3-319-11751-5

VL - 8753

SP - 3

EP - 14

BT - Pattern Recognition: 36th German Conference, GCPR 2014, M{\"u}nster, Germany, September 2-5, 2014, Proceedings

PB - Springer International Publishing AG

ER -