Stochastic Blind Motion Deblurring

Handle URI:
http://hdl.handle.net/10754/556419
Title:
Stochastic Blind Motion Deblurring
Authors:
Xiao, Lei ( 0000-0002-3132-9234 ) ; Gregson, James; Heide, Felix; Heidrich, Wolfgang ( 0000-0002-4227-8508 )
Abstract:
Blind motion deblurring from a single image is a highly under-constrained problem with many degenerate solutions. A good approximation of the intrinsic image can therefore only be obtained with the help of prior information in the form of (often non-convex) regularization terms for both the intrinsic image and the kernel. While the best choice of image priors is still a topic of ongoing investigation, this research is made more complicated by the fact that historically each new prior requires the development of a custom optimization method. In this paper, we develop a stochastic optimization method for blind deconvolution. Since this stochastic solver does not require the explicit computation of the gradient of the objective function and uses only efficient local evaluation of the objective, new priors can be implemented and tested very quickly. We demonstrate that this framework, in combination with different image priors produces results with PSNR values that match or exceed the results obtained by much more complex state-of-the-art blind motion deblurring algorithms.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Citation:
Stochastic Blind Motion Deblurring 2015:1 IEEE Transactions on Image Processing
Publisher:
Institute of Electrical and Electronics Engineers (IEEE)
Journal:
IEEE Transactions on Image Processing
Issue Date:
13-May-2015
DOI:
10.1109/TIP.2015.2432716
Type:
Article
ISSN:
1057-7149; 1941-0042
Additional Links:
http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7106534
Appears in Collections:
Articles; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorXiao, Leien
dc.contributor.authorGregson, Jamesen
dc.contributor.authorHeide, Felixen
dc.contributor.authorHeidrich, Wolfgangen
dc.date.accessioned2015-06-04T12:02:47Zen
dc.date.available2015-06-04T12:02:47Zen
dc.date.issued2015-05-13en
dc.identifier.citationStochastic Blind Motion Deblurring 2015:1 IEEE Transactions on Image Processingen
dc.identifier.issn1057-7149en
dc.identifier.issn1941-0042en
dc.identifier.doi10.1109/TIP.2015.2432716en
dc.identifier.urihttp://hdl.handle.net/10754/556419en
dc.description.abstractBlind motion deblurring from a single image is a highly under-constrained problem with many degenerate solutions. A good approximation of the intrinsic image can therefore only be obtained with the help of prior information in the form of (often non-convex) regularization terms for both the intrinsic image and the kernel. While the best choice of image priors is still a topic of ongoing investigation, this research is made more complicated by the fact that historically each new prior requires the development of a custom optimization method. In this paper, we develop a stochastic optimization method for blind deconvolution. Since this stochastic solver does not require the explicit computation of the gradient of the objective function and uses only efficient local evaluation of the objective, new priors can be implemented and tested very quickly. We demonstrate that this framework, in combination with different image priors produces results with PSNR values that match or exceed the results obtained by much more complex state-of-the-art blind motion deblurring algorithms.en
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en
dc.relation.urlhttp://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7106534en
dc.rights(c) 2015 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.en
dc.titleStochastic Blind Motion Deblurringen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journalIEEE Transactions on Image Processingen
dc.eprint.versionPost-printen
dc.contributor.institutionUniversity of British Columbia, Department of Computer Science, Vancouver, V6T1Z4, Canadaen
kaust.authorHeidrich, Wolfgangen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.