Object dependent manifold priors for image deconvolution

Jie Ni, Pavan Turaga, Vishal M. Patel, Rama Chellappa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Deblurring is an inverse problem which has traditionally been studied from a signal processing perspective. In this paper we consider the role of extra information in the form of prior knowledge of the object class to solve this problem. Specifically, we incorporate unlabeled image data of the object class, say natural images, in the form of a patch-manifold prior for the object class. The manifold is implicitly estimated from the given unlabeled data. We show how the patch manifold prior effectively exploits the availability of the sample class data for regularizing the deblurring problem.

Original languageEnglish (US)
Title of host publicationOptics InfoBase Conference Papers
StatePublished - 2010
Externally publishedYes
EventDigital Image Processing and Analysis, DIPA 2010 - Tucson, AZ, United States
Duration: Jun 7 2010Jun 8 2010

Other

OtherDigital Image Processing and Analysis, DIPA 2010
CountryUnited States
CityTucson, AZ
Period6/7/106/8/10

Fingerprint

Deconvolution
Inverse problems
Signal processing
Availability
availability
signal processing

ASJC Scopus subject areas

  • Instrumentation
  • Atomic and Molecular Physics, and Optics

Cite this

Ni, J., Turaga, P., Patel, V. M., & Chellappa, R. (2010). Object dependent manifold priors for image deconvolution. In Optics InfoBase Conference Papers

Object dependent manifold priors for image deconvolution. / Ni, Jie; Turaga, Pavan; Patel, Vishal M.; Chellappa, Rama.

Optics InfoBase Conference Papers. 2010.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Ni, J, Turaga, P, Patel, VM & Chellappa, R 2010, Object dependent manifold priors for image deconvolution. in Optics InfoBase Conference Papers. Digital Image Processing and Analysis, DIPA 2010, Tucson, AZ, United States, 6/7/10.
Ni J, Turaga P, Patel VM, Chellappa R. Object dependent manifold priors for image deconvolution. In Optics InfoBase Conference Papers. 2010
Ni, Jie ; Turaga, Pavan ; Patel, Vishal M. ; Chellappa, Rama. / Object dependent manifold priors for image deconvolution. Optics InfoBase Conference Papers. 2010.
@inproceedings{1e441cff0e414df3a1b093912b83042a,
title = "Object dependent manifold priors for image deconvolution",
abstract = "Deblurring is an inverse problem which has traditionally been studied from a signal processing perspective. In this paper we consider the role of extra information in the form of prior knowledge of the object class to solve this problem. Specifically, we incorporate unlabeled image data of the object class, say natural images, in the form of a patch-manifold prior for the object class. The manifold is implicitly estimated from the given unlabeled data. We show how the patch manifold prior effectively exploits the availability of the sample class data for regularizing the deblurring problem.",
author = "Jie Ni and Pavan Turaga and Patel, {Vishal M.} and Rama Chellappa",
year = "2010",
language = "English (US)",
isbn = "9781557528926",
booktitle = "Optics InfoBase Conference Papers",

}

TY - GEN

T1 - Object dependent manifold priors for image deconvolution

AU - Ni, Jie

AU - Turaga, Pavan

AU - Patel, Vishal M.

AU - Chellappa, Rama

PY - 2010

Y1 - 2010

N2 - Deblurring is an inverse problem which has traditionally been studied from a signal processing perspective. In this paper we consider the role of extra information in the form of prior knowledge of the object class to solve this problem. Specifically, we incorporate unlabeled image data of the object class, say natural images, in the form of a patch-manifold prior for the object class. The manifold is implicitly estimated from the given unlabeled data. We show how the patch manifold prior effectively exploits the availability of the sample class data for regularizing the deblurring problem.

AB - Deblurring is an inverse problem which has traditionally been studied from a signal processing perspective. In this paper we consider the role of extra information in the form of prior knowledge of the object class to solve this problem. Specifically, we incorporate unlabeled image data of the object class, say natural images, in the form of a patch-manifold prior for the object class. The manifold is implicitly estimated from the given unlabeled data. We show how the patch manifold prior effectively exploits the availability of the sample class data for regularizing the deblurring problem.

UR - http://www.scopus.com/inward/record.url?scp=84896753284&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84896753284&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84896753284

SN - 9781557528926

BT - Optics InfoBase Conference Papers

ER -