Master 2014 2015
Stages de la spécialité SAR
Movement modeling using generative probabilistic models : cases studies

Lieu :Ircam, équipe ISMM, 1, place Igor Stravinsky, 75004 Paris
Encadrant : Frédéric Bevilacqua et Jules Françoise, Ircam, STMS,
Dates :16/02/2015-15/07/2015 (5mois)
Rémunération :500 euros par mois
Mots-clés : Parcours ATIAM : Informatique musicale, Parcours ATIAM : Traitement du signal


Modeling movement is a challenging task that is necessary for the design of gesture-based interfaces. First, the continuous aspect of the movement must be taken into account with high temporal precision. Second, the movement variations must also be quantified since such variations can be related to expressive features. While probabilistic models, such as Hidden Markov Models, have been used extensively for recognition tasks, they have been less explored for movement modeling per se. Recently, several groups have successfully shown that such probabilistic models represent powerful tools for movement modeling and movement generation, but might necessitate to be parameterized differently. Recently Jules Françoise et al proposed a generic framework of probabilistic models [Françoise et al, NIME 2014]. The goal of the proposed internship is to explore these models in selected case studies, from musician to dancer movements. A selection of movement databases will be chosen to evaluate these models. In particular, the possibility to use these models to generate new data will be explored.


Françoise, J., Schnell, N., Borghesi, R., and Bevilacqua, F. (2014). Probabilistic Models for Designing Motion and Sound Relationships. In Proceedings of the International Conference on New Interfaces for Musical Expression (NIME’14), London, UK.

Bevilacqua, F., SchnellN., Rasamimanana N., Zamborlin B., Guédy F. Online Gesture Analysis and Control of Audio Processing, Musical Robots and Interactive Multimodal Systems. Springer Tracts in Advanced Robotics Vol 74, ed. Jorge Solis and Kia C. Ng (Springer Verlag), 2011.