Sublabel-Accurate Convex Relaxation with Total Generalized Variation Regularization

Strecke, Michael; Goldlücke, Bastian

Sublabel-Accurate Convex Relaxation with Total Generalized Variation Regularization

Dateien

Zu diesem Dokument gibt es keine Dateien.

Datum

2019

Autor:innen

Strecke, Michael

Goldlücke, Bastian

DOI (zitierfähiger Link)

https://doi.org/10.1007/978-3-030-12939-2_19

EU-Projektnummer

336978

Projekt

LIA - Light Field Imaging and Analysis

Sammlungen

Informatik und Informationswissenschaft

Publikationstyp

Beitrag zu einem Konferenzband

Publikationsstatus

Published

Erschienen in

BROX, Thomas, ed., Andrés BRUHN, ed., Mario FRITZ, ed.. Pattern Recognition. Cham: Springer Nature Switzerland AG, 2019, pp. 263-277. ISBN 978-3-030-12938-5. Available under: doi: 10.1007/978-3-030-12939-2_19

Zusammenfassung

We propose a novel idea to introduce regularization based on second order total generalized variation (TGV) into optimization frameworks based on functional lifting. The proposed formulation extends a recent sublabel-accurate relaxation for multi-label problems and thus allows for accurate solutions using only a small number of labels, significantly improving over previous approaches towards lifting the total generalized variation. Moreover, even recent sublabel accurate methods exhibit staircasing artifacts when used in conjunction with common first order regularizers such as the total variation (TV). This becomes very obvious for example when computing derivatives of disparity maps computed with these methods to obtain normals, which immediately reveals their local flatness and yields inaccurate normal maps. We show that our approach is effective in reducing these artifacts, obtaining disparity maps with a smooth normal field in a single optimization pass.

Fachgebiet (DDC)

004 Informatik

Konferenz

40th German Conference, GCPR 2018, 9. Okt. 2018 - 12. Okt. 2018, Stuttgart

Zitieren

ISO 690

STRECKE, Michael, Bastian GOLDLÜCKE, 2019. Sublabel-Accurate Convex Relaxation with Total Generalized Variation Regularization. 40th German Conference, GCPR 2018. Stuttgart, 9. Okt. 2018 - 12. Okt. 2018. In: BROX, Thomas, ed., Andrés BRUHN, ed., Mario FRITZ, ed.. Pattern Recognition. Cham: Springer Nature Switzerland AG, 2019, pp. 263-277. ISBN 978-3-030-12938-5. Available under: doi: 10.1007/978-3-030-12939-2_19

BibTex

@inproceedings{Strecke2019Subla-57128,
  year={2019},
  doi={10.1007/978-3-030-12939-2_19},
  title={Sublabel-Accurate Convex Relaxation with Total Generalized Variation Regularization},
  isbn={978-3-030-12938-5},
  publisher={Springer Nature Switzerland AG},
  address={Cham},
  booktitle={Pattern Recognition},
  pages={263--277},
  editor={Brox, Thomas and Bruhn, Andrés and Fritz, Mario},
  author={Strecke, Michael 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/123456789/57128">
    <dc:contributor>Strecke, Michael</dc:contributor>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2022-03-31T14:07:21Z</dc:date>
    <dc:contributor>Goldlücke, Bastian</dc:contributor>
    <dc:creator>Strecke, Michael</dc:creator>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2022-03-31T14:07:21Z</dcterms:available>
    <dcterms:issued>2019</dcterms:issued>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <dc:language>eng</dc:language>
    <dcterms:rights rdf:resource="https://rightsstatements.org/page/InC/1.0/"/>
    <dcterms:title>Sublabel-Accurate Convex Relaxation with Total Generalized Variation Regularization</dcterms:title>
    <dcterms:abstract xml:lang="eng">We propose a novel idea to introduce regularization based on second order total generalized variation (TGV) into optimization frameworks based on functional lifting. The proposed formulation extends a recent sublabel-accurate relaxation for multi-label problems and thus allows for accurate solutions using only a small number of labels, significantly improving over previous approaches towards lifting the total generalized variation. Moreover, even recent sublabel accurate methods exhibit staircasing artifacts when used in conjunction with common first order regularizers such as the total variation (TV). This becomes very obvious for example when computing derivatives of disparity maps computed with these methods to obtain normals, which immediately reveals their local flatness and yields inaccurate normal maps. We show that our approach is effective in reducing these artifacts, obtaining disparity maps with a smooth normal field in a single optimization pass.</dcterms:abstract>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:rights>terms-of-use</dc:rights>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:creator>Goldlücke, Bastian</dc:creator>
    <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/57128"/>
  </rdf:Description>
</rdf:RDF>

Universitätsbibliographie

Ja