TY - GEN
T1 - Universal priors for sparse modeling
AU - Raḿrez, Ignacio
AU - Lecumberry, Federico
AU - Sapiro, Guillermo
PY - 2009
Y1 - 2009
N2 - Sparse data models, where data is assumed to be well represented as a linear combination of a few elements from a dictionary, have gained considerable attention in recent years, and their use has led to state-of-the-art results in many signal and image processing tasks. It is now well understood that the choice of the sparsity regularization term is critical in the success of such models. In this work, we use tools from information theory to propose a sparsity regularization term which has several theoretical and practical advantages over the more standard ℓ0 or ℓ1 ones, and which leads to improved coding performance and accuracy in reconstruction tasks. We also briefly report on further improvements obtained by imposing low mutual coherence and Gram matrix norm on the learned dictionaries.
AB - Sparse data models, where data is assumed to be well represented as a linear combination of a few elements from a dictionary, have gained considerable attention in recent years, and their use has led to state-of-the-art results in many signal and image processing tasks. It is now well understood that the choice of the sparsity regularization term is critical in the success of such models. In this work, we use tools from information theory to propose a sparsity regularization term which has several theoretical and practical advantages over the more standard ℓ0 or ℓ1 ones, and which leads to improved coding performance and accuracy in reconstruction tasks. We also briefly report on further improvements obtained by imposing low mutual coherence and Gram matrix norm on the learned dictionaries.
UR - http://www.scopus.com/inward/record.url?scp=77951108336&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77951108336&partnerID=8YFLogxK
U2 - 10.1109/CAMSAP.2009.5413302
DO - 10.1109/CAMSAP.2009.5413302
M3 - Conference contribution
AN - SCOPUS:77951108336
SN - 9781424451807
T3 - CAMSAP 2009 - 2009 3rd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing
SP - 197
EP - 200
BT - CAMSAP 2009 - 2009 3rd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing
T2 - 2009 3rd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2009
Y2 - 13 December 2009 through 16 December 2009
ER -