An Epipolar Volume Autoencoder With Adversarial Loss for Deep Light Field Super-Resolution

Zhu, Minchen; Alperovich, Anna; Johannsen, Ole; Sulc, Antonin; Goldlücke, Bastian

An Epipolar Volume Autoencoder With Adversarial Loss for Deep Light Field Super-Resolution

Files

There are no files associated with this item.

Date

2019

Authors

Zhu, Minchen

DOI (citable link)

10.1109/CVPRW.2019.00236

EU project number

336978

Project

LIA - Light Field Imaging and Analysis

Collections

Informatik und Informationswissenschaft

Publication type

Contribution to a conference collection

Publication status

Published

Published in

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition workshops : CVPRW 2019 : proceedings : 16-20 June 2019, Long Beach, California. - Piscataway, NJ : IEEE, 2019. - pp. 1853-1861. - ISBN 978-1-72812-506-0

Abstract

When capturing a light field of a scene, one typically faces a trade-off between more spatial or more angular resolution. Fortunately, light fields are also a rich source of information for solving the problem of super-resolution. Contrary to single image approaches, where high-frequency content has to be hallucinated to be the most likely source of the downscaled version, sub-aperture views from the light field can help with an actual reconstruction of those details that have been removed by downsampling. In this paper, we propose a three-dimensional generative adversarial autoencoder network to recover the high-resolution light field from a low-resolution light field with a sparse set of viewpoints. We require only three views along both horizontal and vertical axis to increase angular resolution by a factor of three while at the same time increasing spatial resolution by a factor of either two or four in each direction, respectively.

Subject (DDC)

004 Computer Science

Conference

CVPRW 2019, Jun 16, 2019 - Jun 20, 2019, Long Beach, California

Cite This

ISO 690

ZHU, Minchen, Anna ALPEROVICH, Ole JOHANNSEN, Antonin SULC, Bastian GOLDLÜCKE, 2019. An Epipolar Volume Autoencoder With Adversarial Loss for Deep Light Field Super-Resolution. CVPRW 2019. Long Beach, California, Jun 16, 2019 - Jun 20, 2019. In: 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition workshops : CVPRW 2019 : proceedings : 16-20 June 2019, Long Beach, California. Piscataway, NJ:IEEE, pp. 1853-1861. ISBN 978-1-72812-506-0. Available under: doi: 10.1109/CVPRW.2019.00236

BibTex

@inproceedings{Zhu2019-06Epipo-51260,
  year={2019},
  doi={10.1109/CVPRW.2019.00236},
  title={An Epipolar Volume Autoencoder With Adversarial Loss for Deep Light Field Super-Resolution},
  isbn={978-1-72812-506-0},
  publisher={IEEE},
  address={Piscataway, NJ},
  booktitle={2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition workshops : CVPRW 2019 : proceedings : 16-20 June 2019, Long Beach, California},
  pages={1853--1861},
  author={Zhu, Minchen and Alperovich, Anna and Johannsen, Ole and Sulc, Antonin and Goldlücke, Bastian}
}

RDF

<rdf:RDF
    xmlns:dcterms="http://purl.org/dc/terms/"
    xmlns:dc="http://purl.org/dc/elements/1.1/"
    xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    xmlns:bibo="http://purl.org/ontology/bibo/"
    xmlns:dspace="http://digital-repositories.org/ontologies/dspace/0.1.0#"
    xmlns:foaf="http://xmlns.com/foaf/0.1/"
    xmlns:void="http://rdfs.org/ns/void#"
    xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > 
  <rdf:Description rdf:about="https://kops.uni-konstanz.de/server/rdf/resource/123456789/51260">
    <dc:contributor>Alperovich, Anna</dc:contributor>
    <dc:creator>Alperovich, Anna</dc:creator>
    <dcterms:abstract xml:lang="eng">When capturing a light field of a scene, one typically faces a trade-off between more spatial or more angular resolution. Fortunately, light fields are also a rich source of information for solving the problem of super-resolution. Contrary to single image approaches, where high-frequency content has to be hallucinated to be the most likely source of the downscaled version, sub-aperture views from the light field can help with an actual reconstruction of those details that have been removed by downsampling. In this paper, we propose a three-dimensional generative adversarial autoencoder network to recover the high-resolution light field from a low-resolution light field with a sparse set of viewpoints. We require only three views along both horizontal and vertical axis to increase angular resolution by a factor of three while at the same time increasing spatial resolution by a factor of either two or four in each direction, respectively.</dcterms:abstract>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dcterms:issued>2019-06</dcterms:issued>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:creator>Johannsen, Ole</dc:creator>
    <dcterms:title>An Epipolar Volume Autoencoder With Adversarial Loss for Deep Light Field Super-Resolution</dcterms:title>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2020-10-08T08:55:54Z</dcterms:available>
    <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/51260"/>
    <dc:contributor>Johannsen, Ole</dc:contributor>
    <dc:creator>Sulc, Antonin</dc:creator>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2020-10-08T08:55:54Z</dc:date>
    <dc:creator>Goldlücke, Bastian</dc:creator>
    <dc:contributor>Sulc, Antonin</dc:contributor>
    <dc:language>eng</dc:language>
    <dc:contributor>Zhu, Minchen</dc:contributor>
    <dc:contributor>Goldlücke, Bastian</dc:contributor>
    <dc:creator>Zhu, Minchen</dc:creator>
  </rdf:Description>
</rdf:RDF>

Bibliography of Konstanz

Yes