Presentation

PANAMA (Parsimony and New Algorithms for Audio & Signal Modeling) is a joint project-team between Inria and CNRS. It is part of the D5 department (Digital Signals and Images, Robotics) of IRISA.

Building upon the rare scientific culture of the former METISS project-team, at the interface between audio modeling and mathematical signal processing, the global objective of the PANAMA project-team is to develop mathematically founded and algorithmically efficient techniques to model, acquire and process high-dimensional signals, with a strong emphasis on acoustic data. Applications fuel the proposed mathematical and statistical frameworks with practical scenarii, and the developed algorithms are extensively tested on targeted applications. PANAMA’s methodology relies on a closed loop between theoretical investigations, algorithmic development and empirical studies.

Panama2

 

Keywords:

sparse modeling; audio and music signal processing; source separation; machine learning; inverse problems

Application domains

Audio signal processing and modeling are  a central theme of the PANAMA project-team, with strong and growing industrial connections in the field. However, given the many potential applications beyond audio of the models, methods and algorithms that PANAMA develops, applications to other types of signals may be considered, in particular to biomedical signals. These applications …

Research axes

PANAMA’s research activity is organized around three major research axes. A common thread of these three research axes is the need to design low-dimensional structured data models, to adjust the models to real data through learning and discovery, and to develop efficient algorithms to exploit these models on various tasks ranging from low-level signal acquisition …

Scientific foundations

The scientific foundations of PANAMA are focused on sparse representations and probabilistic modeling, and its scientific scope is extended in three major directions beyond what was investigated in METISS: The extension of the sparse representation paradigm towards that of “sparse modeling”, with the challenge of establishing, strengthening and clarifying connections between sparse representations and machine …

Activity reports

Overall positioning At the interface between audio modeling and mathematical signal processing, the global objective of PANAMA is to develop mathematically founded and algorithmically efficient techniques to model, acquire and process high-dimensional signals, with a strong emphasis on acoustic data. Applications fuel the proposed mathematical and statistical frameworks with practical scenarii, and the developed algorithms …

International and industrial relations

In the context of international collaborative projects, long term international academic collaborations have been established with EPFL (Switzerland), The Technion (Israel),  University of Edinburgh, University of Surrey and Queen Mary, University of London (UK). The project team has regular industrial collaborations through CIFRE Ph.D. contracts with Orange Labs and Technicolor, and through bilateral and collaborative …