N. Seichepine's research while affiliated with Institut Mines-Télécom and other places
What is this page?
This page lists the scientific contributions of an author, who either does not have a ResearchGate profile, or has not yet added these contributions to their profile.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
It was automatically created by ResearchGate to create a record of this author's body of work. We create such pages to advance our goal of creating and maintaining the most comprehensive scientific repository possible. In doing so, we process publicly available (personal) data relating to the author as a member of the scientific community.
If you're a ResearchGate member, you can follow this page to keep up with this author's work.
If you are this author, and you don't want us to display this page anymore, please let us know.
Publications (5)
This work introduces a new framework for nonnegative matrix factorization (NMF) in multisensor or multimodal data configurations, where taking into account the mutual dependence that exists between the related parallel streams of data is expected to improve performance. In contrast with previous works that focused on co-factorization methods -where...
In this paper we propose a non-negative matrix factorization (NMF) model with piecewise-constant activation coefficients. This structure is enforced using a total variation penalty on the rows of the activation matrix. The resulting optimization problem is solved with a majorization-minimization procedure. The proposed algorithm is well suited to a...
This paper presents a new method for bimodal nonnegative matrix factorization (NMF). This method is well-suited to situations where two streams of data are concurrently analyzed and are expected to be related by loosely common factors. It allows for a soft co-factorization, which takes into account the relationship that exists between the modalitie...
Starting with a comparison of state of the art criteria and the influence of filtering, this paper presents an analysis tool for multi-temporal SAR images. It is based on the detection of a step change pattern with a generalized maximum likelihood ratio test. Compared to previous works on similar subject, the proposed approach takes into account a...
Citations
... We use the stress to compare the performance of NCF approaches to the Non-negative Matrix Factorization (NMF) ones [5]. To this end, we consider two NMF algorithms: the alternated projected gradient (APG) [9] and the NeNMF [10] algorithm that uses a Nesterov gradient. NMF is used with a view to highlight clusters of delay trajectories in the lines of the right matrix of the factorization, hence the positivity constraint for the factorization. ...
... In order to alleviate such problems caused in the modeling by this constraint of hard coupling, a "softer" assumption of similarity (or with similar properties), that is, not necessarily a strong equality, can be adopted instead (Seichepine, Essid, Fevotte, & Cappe, 2014;Farias, Cohen, & Comon, 2016). Furthermore, different methods can be used to account for a possible misspecification of the HRF. ...
... Des méthodologies de séparation de sources audio en exploitant la NMF multimodale sont proposées dans [SEICHEPINE et al., 2013 ;SOUVIRAÀ-LABASTIE et al., 2015]. Pour l'identification d'un locuteur durant une discussion, les auteurs de [SEICHEPINE et al., 2013] ont proposé de coupler les modalités audio et vidéo et une variante de l'algorithme de la NMF qui a pour but de minimiser la distance entre les signaux d'activations des deux modalités. ...