File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Conference Paper: Wavelet-based hybrid multilinear models for multidimensional image approximation

TitleWavelet-based hybrid multilinear models for multidimensional image approximation
Authors
KeywordsAdaptive Bases
Hybrid Multilinear Models
Multiscale Analysis
Tensor Ensemble Approximation
Wavelet Transform
Issue Date2008
PublisherI E E E. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000349
Citation
Proceedings - International Conference On Image Processing, Icip, 2008, p. 2828-2831 How to Cite?
AbstractThe wavelet transform hierarchically decomposes images with prescribed bases, while multilineal models search for optimal bases to adapt visual data. In this paper, we integrate these two approaches to compactly represent 2D images and 3D volume data. Once a wavelet (packet) decomposition has been performed, the coefficients are subdivided into small blocks most of which have small energy and are pruned. Surviving blocks usually exhibit strong redundancy among different channels and subbands. To exploit this property, we organize the surviving blocks into small tensors, group the tensors into clusters using an EM algorithm, and compactly approximate each cluster using tensor ensemble approximation. Experimental results on images and medical volume data indicate that our approach achieves better approximation quality than wavelet (packet) transforms. © 2008 IEEE.
Persistent Identifierhttp://hdl.handle.net/10722/151949
ISSN
References

 

DC FieldValueLanguage
dc.contributor.authorWu, Qen_US
dc.contributor.authorChen, Cen_US
dc.contributor.authorYu, Yen_US
dc.date.accessioned2012-06-26T06:31:23Z-
dc.date.available2012-06-26T06:31:23Z-
dc.date.issued2008en_US
dc.identifier.citationProceedings - International Conference On Image Processing, Icip, 2008, p. 2828-2831en_US
dc.identifier.issn1522-4880en_US
dc.identifier.urihttp://hdl.handle.net/10722/151949-
dc.description.abstractThe wavelet transform hierarchically decomposes images with prescribed bases, while multilineal models search for optimal bases to adapt visual data. In this paper, we integrate these two approaches to compactly represent 2D images and 3D volume data. Once a wavelet (packet) decomposition has been performed, the coefficients are subdivided into small blocks most of which have small energy and are pruned. Surviving blocks usually exhibit strong redundancy among different channels and subbands. To exploit this property, we organize the surviving blocks into small tensors, group the tensors into clusters using an EM algorithm, and compactly approximate each cluster using tensor ensemble approximation. Experimental results on images and medical volume data indicate that our approach achieves better approximation quality than wavelet (packet) transforms. © 2008 IEEE.en_US
dc.languageengen_US
dc.publisherI E E E. The Journal's web site is located at http://ieeexplore.ieee.org/xpl/conhome.jsp?punumber=1000349en_US
dc.relation.ispartofProceedings - International Conference on Image Processing, ICIPen_US
dc.subjectAdaptive Basesen_US
dc.subjectHybrid Multilinear Modelsen_US
dc.subjectMultiscale Analysisen_US
dc.subjectTensor Ensemble Approximationen_US
dc.subjectWavelet Transformen_US
dc.titleWavelet-based hybrid multilinear models for multidimensional image approximationen_US
dc.typeConference_Paperen_US
dc.identifier.emailYu, Y:yzyu@cs.hku.hken_US
dc.identifier.authorityYu, Y=rp01415en_US
dc.description.naturelink_to_subscribed_fulltexten_US
dc.identifier.doi10.1109/ICIP.2008.4712383en_US
dc.identifier.scopuseid_2-s2.0-69949120475en_US
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-69949120475&selection=ref&src=s&origin=recordpageen_US
dc.identifier.spage2828en_US
dc.identifier.epage2831en_US
dc.publisher.placeUnited Statesen_US
dc.identifier.scopusauthoridWu, Q=51964899100en_US
dc.identifier.scopusauthoridChen, C=9333688600en_US
dc.identifier.scopusauthoridYu, Y=8554163500en_US

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats