Learning Dictionaries as a sum of Kronecker products

Abstract : The choice of an appropriate frame, or dictionary, is a crucial step in the sparse representation of a given class of signals. Traditional dictionary learning techniques generally lead to unstructured dictionaries which are costly to deploy and train, and do not scale well to higher dimensional signals. In order to overcome such limitation, we propose a learning algorithm that constrains the dictionary to be a sum of Kronecker products of smaller sub-dictionaries. This approach, named SuKro, is demonstrated experimentally on an image denoising application.
Document type :
Journal articles
Liste complète des métadonnées

Cited literature [21 references]  Display  Hide  Download

https://hal.inria.fr/hal-01672349
Contributor : Cassio Dantas <>
Submitted on : Sunday, December 24, 2017 - 2:18:55 PM
Last modification on : Monday, February 11, 2019 - 5:21:32 PM

File

FINAL VERSION.pdf
Files produced by the author(s)

Identifiers

Citation

Cassio Fraga Dantas, Michele N. Da Costa, Renato Da Rocha Lopes. Learning Dictionaries as a sum of Kronecker products. IEEE Signal Processing Letters, Institute of Electrical and Electronics Engineers, 2017, 24 (5), pp.559 - 563. ⟨10.1109/LSP.2017.2681159⟩. ⟨hal-01672349⟩

Share

Metrics

Record views

360

Files downloads

105