From Convolutional Sparse Coding To *-NMF Factorization of Time-Frequency Coefficients - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2024

From Convolutional Sparse Coding To *-NMF Factorization of Time-Frequency Coefficients

Matthieu Kowalski

Résumé

Convolutional Dictionary Learning (CDL) is a dictionary learning technique exploiting the translation invariance of elementary signals. In the time-frequency domain, the repetition of elementary frequency patterns can be exploited through the "nonnegative matrix factorization" (NMF) decompositions and extensions, such as semi or complex-NMF, of the spectrogram. We study the links between these two approaches here, and we show in particular that a signal which admits a Convolutive Sparse Coding decomposition admits time-frequency synthesis coefficients that can be decomposed in semi-NMF or complex-NMF. The different approaches are then compared experimentally on synthetic signals.
Fichier principal
Vignette du fichier
ICASSP24_CSC_NMF-2.pdf (842.91 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04608586 , version 1 (13-06-2024)

Identifiants

Citer

Jean-Baptiste Malagnoux, Matthieu Kowalski. From Convolutional Sparse Coding To *-NMF Factorization of Time-Frequency Coefficients. ICASSP 2024 - IEEE International Conference on Acoustics, Speech and Signal Processing, Apr 2024, Séoul, South Korea. pp.5530-5534, ⟨10.1109/ICASSP48485.2024.10447466⟩. ⟨hal-04608586⟩
0 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Mastodon Facebook X LinkedIn More