A Superresolution Framework for High-Accuracy Multiview Reconstruction


Dateien zu dieser Ressource

Dateien Größe Format Anzeige

Zu diesem Dokument gibt es keine Dateien.

GOLDLÜCKE, Bastian, Mathieu AUBRY, Kalin KOLEV, Daniel CREMERS, 2014. A Superresolution Framework for High-Accuracy Multiview Reconstruction. In: International Journal of Computer Vision. 106(2), pp. 172-191. ISSN 0920-5691. eISSN 1573-1405

@article{Goldlucke2014Super-29111, title={A Superresolution Framework for High-Accuracy Multiview Reconstruction}, year={2014}, doi={10.1007/s11263-013-0654-8}, number={2}, volume={106}, issn={0920-5691}, journal={International Journal of Computer Vision}, pages={172--191}, author={Goldlücke, Bastian and Aubry, Mathieu and Kolev, Kalin and Cremers, Daniel}, note={received DAGM main prize (best paper award)} }

<rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:bibo="http://purl.org/ontology/bibo/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:dcterms="http://purl.org/dc/terms/" xmlns:xsd="http://www.w3.org/2001/XMLSchema#" > <rdf:Description rdf:about="https://kops.uni-konstanz.de/rdf/resource/123456789/29111"> <dcterms:title>A Superresolution Framework for High-Accuracy Multiview Reconstruction</dcterms:title> <bibo:uri rdf:resource="http://kops.uni-konstanz.de/handle/123456789/29111"/> <dcterms:abstract xml:lang="eng">We present a variational framework to estimate super-resolved texture maps on a 3D geometry model of a surface from multiple images. Given the calibrated images and the reconstructed geometry, the proposed functional is convex in the super-resolution texture. Using a conformal atlas of the surface, we transform the model from the curved geometry to the flat charts and solve it using state-of-the-art and provably convergent primal–dual algorithms. In order to improve image alignment and quality of the texture, we extend the functional to also optimize for a normal displacement map on the surface as well as the camera calibration parameters. Since the sub-problems for displacement and camera parameters are non-convex, we revert to relaxation schemes in order to robustly estimate a minimizer via sequential convex programming. Experimental results confirm that the proposed super-resolution framework allows to recover textured models with significantly higher level-of-detail than the individual input images.</dcterms:abstract> <dcterms:available rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2014-10-14T12:26:35Z</dcterms:available> <dc:creator>Goldlücke, Bastian</dc:creator> <dc:creator>Cremers, Daniel</dc:creator> <dc:date rdf:datatype="http://www.w3.org/2001/XMLSchema#dateTime">2014-10-14T12:26:35Z</dc:date> <dc:contributor>Cremers, Daniel</dc:contributor> <dcterms:issued>2014</dcterms:issued> <dc:contributor>Kolev, Kalin</dc:contributor> <dc:contributor>Aubry, Mathieu</dc:contributor> <dc:creator>Aubry, Mathieu</dc:creator> <dc:contributor>Goldlücke, Bastian</dc:contributor> <dc:language>eng</dc:language> <dc:creator>Kolev, Kalin</dc:creator> </rdf:Description> </rdf:RDF>

Das Dokument erscheint in:

KOPS Suche


Mein Benutzerkonto