National audienceIn this article, we consider the dimension reduction problem in the context of blind source separation. We propose (1) a new model based on approximate joint block diagonalization, which allows to find the sources along with their subspace simultaneously, and (2), a Riemannian optimization approach to solve it, original in this context. Our model recovers sources that are independant from the noise and less sensible to the à priori estimation of the number of sources. The efficiency of our model as compared to the state of the art is illustrated through the source separation of an electroencephalographic recording.Dans cet article, nous considérons le problème de la réduction de dimension dans le cadre de la séparation aveu...
CARDOSO Jean-François, GASSIAT Elisabeth (Rapporteurs) DEGERINE Serge, GAMBOA Fabrice (Examinateurs)...
In this thesis, Blind Source Separation (BSS) of Convolutive Mixtures of Sources is addressed. For s...
Separation of sources is an important problem in signal processing where one tries to extract two o...
National audienceIn this article, we consider the dimension reduction problem in the context of blin...
The approximate joint diagonalisation of a set of matrices allows the solution of the blind source s...
Le problème de la Séparation Aveule de Sources est abordé sous l'angle géométrique. Le modèle de mél...
Separation of independent sources using Blind Source Separation (BSS) techniques requires prior know...
Dans cet article, nous proposons une méthode géométrique simple pour la séparation aveugle de source...
Separation of independent sources using Blind Source Separation (BSS) techniques requires prior know...
Separation of independent sources using Blind Source Separation (BSS) techniques requires prior know...
International audienceWe consider the blind source separation (BSS) problem and the closely related ...
Real blind source separation scenarios are rarely “square ” (have equal number of sources as the num...
La séparation aveugle de source consiste à estimer les signaux de sources uniquement à partir des mé...
Blind Source Separation aim to recover a set of M independent signals called sources from the observ...
International audienceApproximate Joint Diagonalization of a matrix set can solve the linear Blind S...
CARDOSO Jean-François, GASSIAT Elisabeth (Rapporteurs) DEGERINE Serge, GAMBOA Fabrice (Examinateurs)...
In this thesis, Blind Source Separation (BSS) of Convolutive Mixtures of Sources is addressed. For s...
Separation of sources is an important problem in signal processing where one tries to extract two o...
National audienceIn this article, we consider the dimension reduction problem in the context of blin...
The approximate joint diagonalisation of a set of matrices allows the solution of the blind source s...
Le problème de la Séparation Aveule de Sources est abordé sous l'angle géométrique. Le modèle de mél...
Separation of independent sources using Blind Source Separation (BSS) techniques requires prior know...
Dans cet article, nous proposons une méthode géométrique simple pour la séparation aveugle de source...
Separation of independent sources using Blind Source Separation (BSS) techniques requires prior know...
Separation of independent sources using Blind Source Separation (BSS) techniques requires prior know...
International audienceWe consider the blind source separation (BSS) problem and the closely related ...
Real blind source separation scenarios are rarely “square ” (have equal number of sources as the num...
La séparation aveugle de source consiste à estimer les signaux de sources uniquement à partir des mé...
Blind Source Separation aim to recover a set of M independent signals called sources from the observ...
International audienceApproximate Joint Diagonalization of a matrix set can solve the linear Blind S...
CARDOSO Jean-François, GASSIAT Elisabeth (Rapporteurs) DEGERINE Serge, GAMBOA Fabrice (Examinateurs)...
In this thesis, Blind Source Separation (BSS) of Convolutive Mixtures of Sources is addressed. For s...
Separation of sources is an important problem in signal processing where one tries to extract two o...