Math @ Duke

Publications [#264701] of Guillermo Sapiro
Papers Published
 Qiu, Q; Sapiro, G; Qiu, Q; Sapiro, G, Learning compressed image classification featuresLearning compressed image classification features,
2014 IEEE International Conference on Image Processing, ICIP 2014
(January, 2014),
pp. 57615765, ISBN 9781479957514 [doi]
(last updated on 2017/12/17)
Abstract: © 2014 IEEE. Learning a transformationbased dimension reduction, thereby compressive, technique for classification is here proposed. Highdimensional data often approximately lie in a union of lowdimensional subspaces. We propose to perform dimension reduction by learning a 'fat' linear transformation matrix on subspaces using nuclear norm as the optimization criteria. The learned transformation enables dimension reduction, and, at the same time, restores a lowrank structure for data from the same class and maximizes the separation between different classes, thereby improving classification via learned lowdimensional features. Theoretical and experimental results support the proposed framework, which can be interpreted as learning compressing sensing matrices for classification.


dept@math.duke.edu
ph: 919.660.2800
fax: 919.660.2821
 
Mathematics Department
Duke University, Box 90320
Durham, NC 277080320

