|
UMR CNRS 6074
Research topics
DIGITAL IMAGERY
-
BUNRAKU - Perception, decision-making and resulting action of real and virtual humans in virtual environments and impact on real environments :
Production methods for virtual-reality and simulation applications: Exploration of new user-immersion interaction methods using visual and haptic interfaces. Virtual-human integration techniques for motricity and high-level behaviour specification. Software development: OpenMASK (created by the team for distribution as an Open Source software platform for virtual reality; checked by expert research).
-
IMADOC - Interpretation and recognition of images and documents :
Analysis of paper documents, handwritten text recognition and human-machine communication using a “digital” pen.
Industrial applications intended for the general public, in fields such as the consultation of archive documents and mobile telephony (smartphone).
Software development : RESIF (On-line Recognition Systems for Handwritten Text using Fuzzy Logic), DocRead (Automatic generator of recognition systems for structured documents), Consultation and annotation platform for archive documents
MODELLING, DIAGNOSIS AND LEARNING
-
DREAM - Diagnosis, recommendations for action and modelling :
Aid for the monitoring and diagnosis of evolving systems and complex activities. This involves inferring the state of a system using observations obtained from sensors to detect and characterise possible dysfunctions.
-
MODAL - Algebraic modelling of dynamic systems :
« Black box » approach to algebraic modelling, symbolic computation and neural network for evaluation and use of these methods in applications related to the environment and health.
-
SIMA2 - Simulation, modelling and automatic learning :
Use of fuzzy systems in digital knowledge processing with focus on automatic knowledge extraction and reinforcement learning.
INFORMATION SYSTEMS
-
LIS - Logical Information Systems :
Development of formal models for information systems that allow for navigation, querying, updating, and analysis of heterogeneous data collections. Data are given descriptors, which may be formal or not, and which are extracted from the data itself or are provided by an external, possibly subjective, judgment. The base model for the representation of information is logical concept analysis, a variant of formal concept analysis. This model naturally allows for navigation, querying, and analysis of data. The team implemented these principles both as a file system LISFS, and as a user-oriented application CAMELIS.
-
TEXMEX - Efficient Exploitation of Multimedia Documents - Exploration, Indexing, and Access to Very Large Databases.
TexMex's research is organised into three complementary directions with the objective of being able to segment, structure and describe large streams of multimedia content in order to re-combine them and re-use them for domestic or professional applications. Firstly, development of advanced data analysis, description and indexing algorithms to facilitate huge amounts of multimedia data. Secondly, research on privileged ways of carrying high-level semantic information such as new techniques for linguistic information acquisition and usage, and the natural language found in speech on audio tracks, (in textual format or incrusted in images or videos). Thirdly, the building of new processing tools for audiovisual documents and mixing cues from the various media to achieve segmentation and structuring of TV or audio streams.
NETWORKS AND SYSTEMS
-
ARMOR - Network Architectures and Models :
Identification, design and selection of the architectures most suitable for the installation of a communication service and development of computing and mathematical tools to carry out tasks related to this service while taking account of the qualitative and quantitative aspects of communication systems.
-
ASAP - As Scalable As possible :
Foundations of large-scale dynamic distributed systems. ASAP's objective is to provide a set of abstractions and algorithms to build serverless, large-scale, distributed applications involving a large set of volatile, geographically-distant, potentially mobile and/or resource-limited computing entities. The ASAP Project-Team is engaged in research along three main themes: Distributed computing models and abstractions; Peer-to-peer distributed systems and applications; Data management in wireless autonomic networks.
These research activities seek conceptual advances in both basic and applied research, then validation of these concepts through real applications.
-
PARIS - Programming of parallel and distributed systems for large-scale digital simulation :
Programming of grids for applications using distributed digital simulation techniques, with which it is possible to reduce the lead-time in the design of manufactured products (automobiles, aircraft...).
Grids are highly-distributed infrastructures grouping together a very diverse range of machines from PC networks to supercomputers, with the advantage that they provide great computing power as well as a very large capacity for data storage.
Technology transfer
-
Industry : Nexter, France Telecom R&D, Conseil Général des Yvelines, Archives de France
-
Start-up creation : Evodia, Avel Vor Technologie
-
National projects ANR ACI MD : Madone, Demi-Ton, Navidomass, Alpage
-
National projects ANR : Part@ge, Semim@ge, SensLab, SVP, SafeScale, Icos-HD
Equipment
Technical platform :
-
Immersive display system (9-m circumference cylindrical screen, 3 coupled projectors)
-
Haptic interactive device (Virtuoso 6D – Haption arm)
-
Human movement acquisition system (12 VICON infra-red cameras)
-
Grid computer : Grid5000
-
Platform TELEMEX real-time video indexation
Partnerships and international relations
-
NoE : INTUITION, CoreGrid, S-Cube, ReSIST, Muscle
-
European Projects : aceMedia, WS-Diamond
|
|
|
