2.50
Hdl Handle:
http://hdl.handle.net/10547/337005
Title:
Enhancing Bayesian estimators for removing camera shake
Authors:
Wang, Chao; Yue, Y.; Dong, Feng; Tao, Yubo; Ma, Xiangyin; Clapworthy, Gordon J.; Ye, Xujiong
Abstract:
The aim of removing camera shake is to estimate a sharp version x from a shaken image y when the blur kernel k is unknown. Recent research on this topic evolved through two paradigms called MAP(k) and MAP(x,k). MAP(k) only solves for k by marginalizing the image prior, while MAP(x,k) recovers both x and k by selecting the mode of the posterior distribution. This paper first systematically analyses the latent limitations of these two estimators through Bayesian analysis. We explain the reason why it is so difficult for image statistics to solve the previously reported MAP(x,k) failure. Then we show that the leading MAP(x,k) methods, which depend on efficient prediction of large step edges, are not robust to natural images due to the diversity of edges. MAP(k), although much more robust to diverse edges, is constrained by two factors: the prior variation over different images, and the ratio between image size and kernel size. To overcome these limitations, we introduce an inter-scale prior prediction scheme and a principled mechanism for integrating the sharpening filter into MAP(k). Both qualitative results and extensive quantitative comparisons demonstrate that our algorithm outperforms state-of-the-art methods.
Affiliation:
University of Bedfordshire
Citation:
Wang, C., Yue, Y., Dong, F., Tao, Y., Ma, X., Clapworthy, G.J. (2013) 'Enhancing Bayesian Estimators for Removing Camera Shake', Computer Graphics Forum, 32 (6) pp113-125.
Publisher:
Wiley
Journal:
Computer Graphics Forum
Issue Date:
2013
URI:
http://hdl.handle.net/10547/337005
DOI:
10.1111/cgf.12074
Additional Links:
http://doi.wiley.com/10.1111/cgf.12074
Type:
Article
Language:
en
ISSN:
0167-7055
Appears in Collections:
Centre for Computer Graphics and Visualisation (CCGV)

Full metadata record

DC FieldValue Language
dc.contributor.authorWang, Chaoen
dc.contributor.authorYue, Y.en
dc.contributor.authorDong, Fengen
dc.contributor.authorTao, Yuboen
dc.contributor.authorMa, Xiangyinen
dc.contributor.authorClapworthy, Gordon J.en
dc.contributor.authorYe, Xujiongen
dc.date.accessioned2014-12-10T10:47:32Z-
dc.date.available2014-12-10T10:47:32Z-
dc.date.issued2013-
dc.identifier.citationWang, C., Yue, Y., Dong, F., Tao, Y., Ma, X., Clapworthy, G.J. (2013) 'Enhancing Bayesian Estimators for Removing Camera Shake', Computer Graphics Forum, 32 (6) pp113-125.en
dc.identifier.issn0167-7055-
dc.identifier.doi10.1111/cgf.12074-
dc.identifier.urihttp://hdl.handle.net/10547/337005-
dc.description.abstractThe aim of removing camera shake is to estimate a sharp version x from a shaken image y when the blur kernel k is unknown. Recent research on this topic evolved through two paradigms called MAP(k) and MAP(x,k). MAP(k) only solves for k by marginalizing the image prior, while MAP(x,k) recovers both x and k by selecting the mode of the posterior distribution. This paper first systematically analyses the latent limitations of these two estimators through Bayesian analysis. We explain the reason why it is so difficult for image statistics to solve the previously reported MAP(x,k) failure. Then we show that the leading MAP(x,k) methods, which depend on efficient prediction of large step edges, are not robust to natural images due to the diversity of edges. MAP(k), although much more robust to diverse edges, is constrained by two factors: the prior variation over different images, and the ratio between image size and kernel size. To overcome these limitations, we introduce an inter-scale prior prediction scheme and a principled mechanism for integrating the sharpening filter into MAP(k). Both qualitative results and extensive quantitative comparisons demonstrate that our algorithm outperforms state-of-the-art methods.en
dc.language.isoenen
dc.publisherWileyen
dc.relation.urlhttp://doi.wiley.com/10.1111/cgf.12074en
dc.rightsArchived with thanks to Computer Graphics Forumen
dc.subjectblind deconvolutionen
dc.subjectBayesian estimatoren
dc.subjectimage deblurringen
dc.subjectimage processingen
dc.subjectsharpeningen
dc.subjectcomputer visionen
dc.subjectdeblurringen
dc.titleEnhancing Bayesian estimators for removing camera shakeen
dc.typeArticleen
dc.contributor.departmentUniversity of Bedfordshireen
dc.identifier.journalComputer Graphics Forumen
All Items in UOBREP are protected by copyright, with all rights reserved, unless otherwise indicated.