Structure-from-Motion-Aware PatchMatch for Adaptive Optical Flow Estimation

Lade...
Vorschaubild
Dateien
Zu diesem Dokument gibt es keine Dateien.
Datum
2018
Autor:innen
Maurer, Daniel
Bruhn, Andrés
Herausgeber:innen
Kontakt
ISSN der Zeitschrift
Electronic ISSN
ISBN
Bibliografische Daten
Verlag
Schriftenreihe
Auflagebezeichnung
URI (zitierfähiger Link)
ArXiv-ID
Internationale Patentnummer
EU-Projektnummer
DFG-Projektnummer
Angaben zur Forschungsförderung (Freitext)
Projekt
Open Access-Veröffentlichung
Gesperrt bis
Titel in einer weiteren Sprache
Forschungsvorhaben
Organisationseinheiten
Zeitschriftenheft
Publikationstyp
Beitrag zu einem Konferenzband
Publikationsstatus
Published
Erschienen in
FERRARI, Vittorio, ed. and others. Computer Vision - ECCV 2018 : 15th European Conference, Munich, Germany, September 8-14, 2018, Proceedings, Part VIII. Cham: Springer, 2018, pp. 575-592. Lecture Notes in Computer Science. 11212. ISSN 0302-9743. eISSN 1611-3349. ISBN 978-3-030-01236-6. Available under: doi: 10.1007/978-3-030-01237-3_35
Zusammenfassung

Many recent energy-based methods for optical flow estimation rely on a good initialization that is typically provided by some kind of feature matching. So far, however, these initial matching approaches are rather general: They do not incorporate any additional information that could help to improve the accuracy or the robustness of the estimation. In particular, they do not exploit potential cues on the camera poses and the thereby induced rigid motion of the scene. In the present paper, we tackle this problem. To this end, we propose a novel structure-from-motion-aware PatchMatch approach that, in contrast to existing matching techniques, combines two hierarchical feature matching methods: a recent two-frame PatchMatch approach for optical flow estimation (general motion) and a specifically tailored three-frame PatchMatch approach for rigid scene reconstruction (SfM). While the motion PatchMatch serves as baseline with good accuracy, the SfM counterpart takes over at occlusions and other regions with insufficient information. Experiments with our novel SfM-aware PatchMatch approach demonstrate its usefulness. They not only show excellent results for all major benchmarks (KITTI 2012/2015, MPI Sintel), but also improvements up to 50% compared to a PatchMatch approach without structure information.

Zusammenfassung in einer weiteren Sprache
Fachgebiet (DDC)
004 Informatik
Schlagwörter
Konferenz
15th European Conference on Computer Vision, ECCV 2018, 8. Sep. 2018 - 14. Sep. 2018, Munich, Germany
Rezension
undefined / . - undefined, undefined
Zitieren
ISO 690MAURER, Daniel, Nico MARNIOK, Bastian GOLDLÜCKE, Andrés BRUHN, 2018. Structure-from-Motion-Aware PatchMatch for Adaptive Optical Flow Estimation. 15th European Conference on Computer Vision, ECCV 2018. Munich, Germany, 8. Sep. 2018 - 14. Sep. 2018. In: FERRARI, Vittorio, ed. and others. Computer Vision - ECCV 2018 : 15th European Conference, Munich, Germany, September 8-14, 2018, Proceedings, Part VIII. Cham: Springer, 2018, pp. 575-592. Lecture Notes in Computer Science. 11212. ISSN 0302-9743. eISSN 1611-3349. ISBN 978-3-030-01236-6. Available under: doi: 10.1007/978-3-030-01237-3_35
BibTex
@inproceedings{Maurer2018Struc-43731,
  year={2018},
  doi={10.1007/978-3-030-01237-3_35},
  title={Structure-from-Motion-Aware PatchMatch for Adaptive Optical Flow Estimation},
  number={11212},
  isbn={978-3-030-01236-6},
  issn={0302-9743},
  publisher={Springer},
  address={Cham},
  series={Lecture Notes in Computer Science},
  booktitle={Computer Vision - ECCV 2018 : 15th European Conference, Munich, Germany, September 8-14, 2018, Proceedings, Part VIII},
  pages={575--592},
  editor={Ferrari, Vittorio},
  author={Maurer, Daniel and Marniok, Nico and Goldlücke, Bastian and Bruhn, Andrés}
}
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/123456789/43731">
    <dc:contributor>Marniok, Nico</dc:contributor>
    <dc:creator>Goldlücke, Bastian</dc:creator>
    <dc:contributor>Goldlücke, Bastian</dc:contributor>
    <foaf:homepage rdf:resource="http://localhost:8080/"/>
    <dspace:isPartOfCollection rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:creator>Maurer, Daniel</dc:creator>
    <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2018-11-08T10:12:21Z</dc:date>
    <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2018-11-08T10:12:21Z</dcterms:available>
    <bibo:uri rdf:resource="https://kops.uni-konstanz.de/handle/123456789/43731"/>
    <dcterms:isPartOf rdf:resource="https://kops.uni-konstanz.de/server/rdf/resource/123456789/36"/>
    <dc:contributor>Bruhn, Andrés</dc:contributor>
    <dc:creator>Bruhn, Andrés</dc:creator>
    <dcterms:issued>2018</dcterms:issued>
    <dc:creator>Marniok, Nico</dc:creator>
    <dcterms:abstract xml:lang="eng">Many recent energy-based methods for optical flow estimation rely on a good initialization that is typically provided by some kind of feature matching. So far, however, these initial matching approaches are rather general: They do not incorporate any additional information that could help to improve the accuracy or the robustness of the estimation. In particular, they do not exploit potential cues on the camera poses and the thereby induced rigid motion of the scene. In the present paper, we tackle this problem. To this end, we propose a novel structure-from-motion-aware PatchMatch approach that, in contrast to existing matching techniques, combines two hierarchical feature matching methods: a recent two-frame PatchMatch approach for optical flow estimation (general motion) and a specifically tailored three-frame PatchMatch approach for rigid scene reconstruction (SfM). While the motion PatchMatch serves as baseline with good accuracy, the SfM counterpart takes over at occlusions and other regions with insufficient information. Experiments with our novel SfM-aware PatchMatch approach demonstrate its usefulness. They not only show excellent results for all major benchmarks (KITTI 2012/2015, MPI Sintel), but also improvements up to 50% compared to a PatchMatch approach without structure information.</dcterms:abstract>
    <dc:language>eng</dc:language>
    <dc:contributor>Maurer, Daniel</dc:contributor>
    <dcterms:title>Structure-from-Motion-Aware PatchMatch for Adaptive Optical Flow Estimation</dcterms:title>
    <void:sparqlEndpoint rdf:resource="http://localhost/fuseki/dspace/sparql"/>
  </rdf:Description>
</rdf:RDF>
Interner Vermerk
xmlui.Submission.submit.DescribeStep.inputForms.label.kops_note_fromSubmitter
Kontakt
URL der Originalveröffentl.
Prüfdatum der URL
Prüfungsdatum der Dissertation
Finanzierungsart
Kommentar zur Publikation
Allianzlizenz
Corresponding Authors der Uni Konstanz vorhanden
Internationale Co-Autor:innen
Universitätsbibliographie
Ja
Begutachtet
Diese Publikation teilen