Current collaborations

National collaborations

ANR SYSCOMM GEO-FLUIDS (2009-2013)

The project Geo-FLUIDS focuses on the specification of tools to analyse geophysical fluid flows from image sequences. Geo-FLUIDS aims at providing image-based methods using physically consistent models to extract meaningful features describing the observed flow and to unveil the dynamical properties of this flow. The main targeted application domains concern Oceanography and Meteorology. The project consortium gathers the INRIA research groups: FLUMINANCE (leader), CLIME, IPSO, and MOISE. The group of the Laboratoire de Météorologie Dynamique located at the ENS Paris, the IFREMER-CERSAT group located at Brest and the METEOFRANCE GMAP group in Toulouse.

ANR COSINUS PREVASSEMBLE (2009-2012)

The main goal of this project is the further development of ensemble methods for meteorology and oceanography, to be used for prediction as well as for assimilation of observations. The main problems studied in this project consists of the assessment of the feasability of ensemble variational assimilation, the precise assessment of what can and cannot be achieved by ensemble estimation with reduced set of samples, and the development of methods for more accurate evaluation of modelling errors. This project involves the ENS-LMD laboratory, the Meteo-France Research center (CNRM) and INRIA (P. Del Moral, F. Le Gland, E. Mémin).

ANR SYSCOMM MSDAG Spatio-temporal Analysis of deformable structures in Meteosat Second Generation images (2009-2012)

This project includes three INRIA groups (CLIME, FLUMINANCE, MOISE) the Research and Teaching Center in Atmospheric Environment of the Ecole Nationale des Ponts et Chaussées (CEREA). It aims at studying multiscale data assimilation techniques. We will focus in particular on questions related to an optimal choice of the resolution of the control variables with respect to the data and background resolution and the model scale available. Multigrid assimilation principles will be also studied and compared to fixed grid appraoch.

International collaborations

STIC AMSUD Modeling voice production (2008-2011)

This three years project aims at studying both, theoretically and experimentally, the phenomena involved in vocal fold oscillations, towards the elaboration of performing vocal production models, for application in voice and speech technology. It gathers partners from Brazil (PUCRS, UFF), Argentina (UBA) and France (LIMSI,ICP, Fluminance).

Associated team HURACAN (2010-2013)

HURACAN is a collaborative group between the University of Buenos Aires and the Fluminance group. This associated team focuses on the analysis and the control of fluid flow from image sequences. The research objectives of this group are organised in two distinct themes. The first one aims at defining and studying visual servoing techniques for the control of fluid flows. Besides, the issue of defining efficient servoing techniques, this axis gathers also research problems related to the definition of velocity measurements and to flow plasma actuators. The second axis is centered on the coupling between large-scale representations of geophysical flow and satellite image data. It aims more precisely at defining directly from image data dynamical terms related to subgrid modeling of the observed flow. The second research topic is centered on the coupling between image data and large scales representations of geophysical flows. It aims more precisely at defining through the image data the subgrid dynamical terms related to the small scales of the observed flow. This axis of study includes issues of coupling data and model defined at different scales, problems of small scales velocity image measurements obeying turbulence phenomenology representations or fitting well to experimental data.

Industrial collaboration

MSR-Inria Video data mining for environmental and social sciences (2009-2013)

This contract takes place in the context of the joint Microsoft-Inria research laboratory. It involves two other Inria project-teams, Willow and Lear. The projects is composed of three fairly disjoint sub-projects. Fluminance contributes to two of them: (i) Mining dynamical remote data with applications in computational ecology and environmental science; (ii) Mining TV broadcasts with applications to sociology. In the first subproject, we aim at combining various low and mid-level video analysis tools (shot detection, camera motion characterization, visual tracking, object recognition, human action recognition) for the analysis and annotation of human actions and interactions in video segments to assist and provide data for studies of consumer trends in commercials, political event coverage in newscasts, and class- and gender-related behavior patterns in situation comedies, for example. In the second one, we aim at designing new tools for the detection of salient changes in multi-temporal satellite images (with application to assessment of natural damages, consequences of climate modifications, and changes caused by human action in urban or natural environments), and, in the longer term, for the detection, identification and tracking of dynamics meteorological events, with application to risk assessment and weather forecast.