Home

portrait_edit_cropWelcome to my professional webpage! I am currently a tenured research scientist within the project-team PANAMA, which means Parsimony & New Algorithms for Audio & Signal Modeling. PANAMA is a joint research team of Inria Rennes (my employer), and CNRS (UMR 6074 – IRISA).

Research interests in brief

The quickest way to present my research is that I develop mathematical models and algorithms to process and understand audio signals recorded by microphones. Examples of technological applications of this field are hearing aids, interactive robots with auditory abilities, restoration and remixing of music recordings, or automatic speech recognition. My research contributions mostly lie within four fields:

  • Audio signal processing : Sound source localization and separation, signal enhancement
  • Machine learning : convex optimization, nonlinear regression, dictionary learning, Bayes’ statistics
  • Robot audition : Speaker localization and egonoise reduction with the robot NAO
  • Image signal processing / Computer vision : Hyperspectral imaging, face pose estimation

Biography

Since January 2016, Antoine Deleforge is a tenured research scientist (Chargé de recherche) with Inria in Rennes, France, within the PANAMA research group. He received the engineering B.Sc. (2008) and M.Sc. (2010) degrees in computer science and mathematics from the school Ensimag (Grenoble, France), as well as the specialized M.Sc. research degree in computer graphics, vision and robotics from the Université Joseph Fourier (Grenoble, France). In November 2013, he received the Ph.D. degree in computer sciences and applied mathematics of the university of Grenoble (France), under the supervision of Radu Horaud (PhD thesis available online). He was then employed as a postdoctoral fellow for two years (2014-2015) at the chair of Multimedia Communications and Signal Processing of the Friedrich-Alexander-University of Erlangen (Germany) with Prof. Walter Kellerman.

In 2014, Antoine Deleforge received the best French PhD award in signal image and vision, jointly delivered every year by the GRETSI, the club EEA and the GdR ISIS. He also received a best journal paper award in 2016 and a best conference paper award 2015. He was the main organizer of two special sessions on learning-based audio scene geometry estimation at the international conferences ICASSP 2016 and LVA/ICA 2017. He was the main coordinator and programmer of the Journée Science et Musique 2016 (Rennes, France), a wide-audience one-day event revealing the science and technologies behind music and sounds, which attracted around 600 visitors.

CV (October 2016).