Consensus Convolutional Sparse Coding

Handle URI:
http://hdl.handle.net/10754/626836
Title:
Consensus Convolutional Sparse Coding
Authors:
Choudhury, Biswarup; Swanson, Robin; Heide, Felix; Wetzstein, Gordon; Heidrich, Wolfgang
Abstract:
Convolutional sparse coding (CSC) is a promising direction for unsupervised learning in computer vision. In contrast to recent supervised methods, CSC allows for convolutional image representations to be learned that are equally useful for high-level vision tasks and low-level image reconstruction and can be applied to a wide range of tasks without problem-specific retraining. Due to their extreme memory requirements, however, existing CSC solvers have so far been limited to low-dimensional problems and datasets using a handful of low-resolution example images at a time. In this paper, we propose a new approach to solving CSC as a consensus optimization problem, which lifts these limitations. By learning CSC features from large-scale image datasets for the first time, we achieve significant quality improvements in a number of imaging tasks. Moreover, the proposed method enables new applications in high-dimensional feature learning that has been intractable using existing CSC methods. This is demonstrated for a variety of reconstruction problems across diverse problem domains, including 3D multispectral demosaicing and 4D light field view synthesis.
KAUST Department:
Visual Computing Center, King Abdullah University of Science and Technology (KAUST)
Publisher:
IEEE
Journal:
IEEE Xplore
Conference/Event name:
International Conference on Computer Vision (ICCV)
Issue Date:
Dec-2017
DOI:
10.1109/ICCV.2017.459
Type:
Conference Paper
Sponsors:
VCC KAUST Baseline Funding, Terman Faculty Fellowship, Intel Compressive Sensing Alliance, National Science Foundation (IIS 1553333), and NSF/Intel Partnership on Visual and Experiential Computing (IIS 1539120).
Additional Links:
http://ieeexplore.ieee.org/document/8237721/
Appears in Collections:
Conference Papers

Full metadata record

DC FieldValue Language
dc.contributor.authorChoudhury, Biswarupen
dc.contributor.authorSwanson, Robinen
dc.contributor.authorHeide, Felixen
dc.contributor.authorWetzstein, Gordonen
dc.contributor.authorHeidrich, Wolfgangen
dc.date.accessioned2018-01-17T13:58:13Z-
dc.date.available2018-01-17T13:58:13Z-
dc.date.issued2017-12-
dc.identifier.doi10.1109/ICCV.2017.459-
dc.identifier.urihttp://hdl.handle.net/10754/626836-
dc.description.abstractConvolutional sparse coding (CSC) is a promising direction for unsupervised learning in computer vision. In contrast to recent supervised methods, CSC allows for convolutional image representations to be learned that are equally useful for high-level vision tasks and low-level image reconstruction and can be applied to a wide range of tasks without problem-specific retraining. Due to their extreme memory requirements, however, existing CSC solvers have so far been limited to low-dimensional problems and datasets using a handful of low-resolution example images at a time. In this paper, we propose a new approach to solving CSC as a consensus optimization problem, which lifts these limitations. By learning CSC features from large-scale image datasets for the first time, we achieve significant quality improvements in a number of imaging tasks. Moreover, the proposed method enables new applications in high-dimensional feature learning that has been intractable using existing CSC methods. This is demonstrated for a variety of reconstruction problems across diverse problem domains, including 3D multispectral demosaicing and 4D light field view synthesis.en
dc.description.sponsorshipVCC KAUST Baseline Funding, Terman Faculty Fellowship, Intel Compressive Sensing Alliance, National Science Foundation (IIS 1553333), and NSF/Intel Partnership on Visual and Experiential Computing (IIS 1539120).en
dc.publisherIEEEen
dc.relation.urlhttp://ieeexplore.ieee.org/document/8237721/en
dc.subjectConvolutional sparse codingen
dc.subjectConvex Optimization, Machine Learning, Big Data, Image Reconstruction, High-dimensional Dataen
dc.titleConsensus Convolutional Sparse Codingen
dc.typeConference Paperen
dc.contributor.departmentVisual Computing Center, King Abdullah University of Science and Technology (KAUST)en
dc.identifier.journalIEEE Xploreen
dc.conference.dateOctober 22-29, 2017en
dc.conference.nameInternational Conference on Computer Vision (ICCV)en
dc.conference.locationVenice, Italyen
dc.eprint.versionPost-printen
dc.contributor.institutionUniversity of Toronto, Stanford Universityen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.