Home

About

Advanced Search

Browse by Discipline

Scientific Societies

E-print Alerts

Add E-prints

E-print Network
FAQHELPSITE MAPCONTACT US


  Advanced Search  

 
Int J Comput Vis (2008) 76: 217229 DOI 10.1007/s11263-007-0053-0
 

Summary: Int J Comput Vis (2008) 76: 217229
DOI 10.1007/s11263-007-0053-0
A Tensor Approximation Approach to Dimensionality
Reduction
Hongcheng Wang Narendra Ahuja
Received: 6 October 2005 / Accepted: 9 March 2007 / Published online: 30 June 2007
Springer Science+Business Media, LLC 2007
Abstract Dimensionality reduction has recently been ex-
tensively studied for computer vision applications. We
present a novel multilinear algebra based approach to re-
duced dimensionality representation of multidimensional
data, such as image ensembles, video sequences and volume
data. Before reducing the dimensionality we do not convert
it into a vector as is done by traditional dimensionality re-
duction techniques like PCA. Our approach works directly
on the multidimensional form of the data (matrix in 2D and
tensor in higher dimensions) to yield what we call a Datum-
as-Is representation. This helps exploit spatio-temporal
redundancies with less information loss than image-as-
vector methods. An efficient rank-R tensor approximation

  

Source: Ahuja, Narendra - Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign

 

Collections: Computer Technologies and Information Sciences; Engineering