INK-SVD: LEARNING INCOHERENT DICTIONARIES FOR SPARSE REPRESENTATIONS

Boris Mailhé (1) , Daniele Barchiesi (1) , Mark D. Plumbley (2)

1 Centre for Digital Music
2 Centre for Digital Music, Queen Mary University of London

This work considers the problem of learning an incoherent dictionary that is both adapted to a set of training data and incoherent so that existing sparse approximation algorithms can recover the sparsest representation. A new decorrelation method is presented that computes a fixed coherence dictionary close to a given dictionary. That step iterates pairwise decorrelations of atoms in the dictionary. Dictionary learn- ing is then performed by adding this decorrelation method as an intermediate step in the K-SVD learning algorithm. The proposed algorithm INK-SVD is tested on musical data and compared to another existing decorrelation method. INK- SVD can compute a dictionary that approximates the training data as well as K-SVD while decreasing the coherence from 0.6 to 0.2.

Mots clés: Sparse coding; Dictionary learning; Coherence; K-SVD

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Boris Mailhé, Daniele Barchiesi, Mark D. Plumbley. INK-SVD: LEARNING INCOHERENT DICTIONARIES FOR SPARSE REPRESENTATIONS. 2011. ⟨hal-00688518⟩