Projects
DYNI
Ecole Recherche Multimodale d'Information, Tech. & Sciences
ANR AVEIR
ANR ANCL
ANR COGNILEGO
IEEE Passive
Contest Campaigns
2nd/6 on recall precision criteria on violent scene detection in Hollywood movies at Affect Event MediaEval 2011
1st @ Multiple Whale Passive Acoustic Detection & Tracking challenge 2005+2009 DEMO
2nd/4 @ ImagEval 2006 Web image retrieval (after Xerox)
4th/11 @ Visual Concept Detection ImageClef 2008 (after Xerox, RWTH, LIP6)
3th/24 with AVEIR run @ Photo Retrieval Diversity ImageClef 2009 (after XEROX, DCU)
5th/13 @ Visual Concept Detection ICPR2010 ImageClef (after CVSSPret, ISIS, IJS, PTECH, MMIS)
High Level Features detection @ TREC VIDEO 2008-09 (IRIM project)
Ester I (Event Detection / Segmentation Task)
Ester II (Named Entity detection Task)
Lessons
L2 Math-Info Graph Theory
Master-Information Theory formulae
Master SIS - lessons in Content Based Information Retreival
Phd Students
S. Tollari (2003-2006 done)
N. Ben Aloui (2006-2009 done)
S. Fraihat (2007-2010 done)
A. Zidouni (2007-2010 done)
F. Benard (2007-2011)
P. Marchart (2009-)
R. Abeille (2010-)
Y. Doh (2011-)
Varia
UMR LSIS
AFCP (Assoc. Francophone Communication Parlee
Master Sciences de L'information & Systemes
Master Dev. et Appl. Plateformes Mobiles
Journees Rochebrune 1992-2004
CiteSeer

Hervé GLOTIN

Professeur - Awarded and member 2011-15 of the Institut universitaire de France for outstanding research (IUF *)
Lab. Sciences de l'Information et des Systèmes - UMR CNRS 7296 (LSIS)
Univ. Sud Toulon Var R229-BP20132-83957 La Garde CEDEX-France-Tel/Fax:+33(0)4 941428 24/97 glotin (at) univ-tln.fr

INFORMATION DYNAMICS & INTEGRATION project (leader) [DYNI]

Bibliography

Link to my selected citations

Scene Understanding

Speech

Bioacoustics

Cognitive Systems

NEWS:

"Cooperative Sparse Representation in Two Opposite Directions for Semi-supervised Image Annotation" to appear in IEEE Transaction in Image Processing 2012 DOI 10.1109/TIP.2012.2197631

Invited discussant at Institut des Hautes Etudes Scientifiques IHES workshop on Mathematical Models of Sound Analysis - 14 and 15 june 2012

Editor of "Soundscape Semiotics - Localisation and Categorisation" open book project 2012

ERMITES 2012 : Summer school on Representation and Decision in Vision

"Stochastic Low-Rank Kernel Learning for Regression" @ ICML 2011

Leader of the ANR project COGNILEGO : from Pixels to Semantics, with UMR LNIA and A2IA

France Television interview on bioacoutics, April 2012, 13 minutes (MP4 60Mo)

Research Interests :

  • Perceptual Artificial Intelligence
  • Multimodal Information Retrieval, Content Analysis & Mining - ANR AVEIR project (LIP6, CLIPS, Paris Telecom, LSIS) - Pole CAP DIGITAL AVEIR
  • Automatic Speech Recognition - ESTER Project
  • Content Based Image Retrieval and Auto-Annotation   Examples DIMATEX poster
  • Biometrics - Project with DCN : CIFRE PHD
  • Bioacoustics - Project PIMC, multiple whale tracking, patented PCT 2009 - Pole Mer
  • Machine Learning (ANN, SVM, ART2, LDA, ALDA...)
  • Information Theory (Shannon, Fisher...)
  • Cognitive Sciences (Neurobiology of perception, reasoning and learning) - TCAN Project 2005-06 + ANR ANCL 2007-09 (LPC, DDL, LNIA, LSIS resp.)

Habilitation thesis : "Recherche d'Information et Perception Robustes pour une Multi-Structuration Audio-Video-Lego ŕ l'échelle", 2007. Jury : Gallinari P., Jauffret C. , Barhen J., Bengio S., Merialdo B., Scholl M., Le Maitre J.

PHD thesis : "Elaboration et comparaison de systemes adaptatifs multi-flux de reconnaissance robuste de la parole : incorporation des indices de voisement et de localisation", 2001 PDF (10Mo for acroread) PS.GZ (7Mo).

Former projects :

Vision : ACI TCAN and ANR ANCL (2005-2009) on neurocomputationnel reading
Speech recognition: audiovisual VIA VOICE with IBM (2001), Cost 249, ESPRIT  projects: RESPITE et SPHEAR .

Doctorant / Phd Student :

  • sept.2003-oct.2006: Direction (avec J. Le Maitre) de la these de S. Tollari : 'Indexation et recherche d'images par fusion d'information textuelles et visuelles'. Maintenant MC au LIP6.
  • fev.2006-janv. 2009: (avec O. Papini) these de N. Ben Aloui : 'Biometrie multimodale par parametrisation et fusion qualitatives' (CIFRE DCN). Maintenant ing. R&D DCNS.
  • fev.2007-24 mars 2010: these de S. Fraihat : 'Representation parcimonieuse temps-frequence de la parole' (projet LSIS). Maintenant enseignant dans le superieur prive sur Paris.
  • oct.2007-2011: these de F. Caudal-Benard : 'Trajectographie temps-reel par acoustique passive et analyse des emissions de mammiferes marins' (REGION+PME 10%).
  • oct.2007-dec 2010: these de A. Zidouni : 'MODELES GRAPHIQUES DISCRIMINANTS pour l'etiquetage de sequences : Application a la Reconnaissance d'Entites Nommees Radiophoniques' (MNRT).Maintenant Ing R. et D. chez Thales.
  • oct.2009-...: (avec Liva Ralaivola) these de P. Machart : 'Cotraining Multimodal' (MNRT).
  • oct.2010-...: (avec Pascale Giraudet) these de R. Abeille : 'Analyse Acoustique sous marine pour modeles de biopopulation de cetaces' (USTV).
  • nov.2011-...: (avec Joseph Razik et Olivier Adam) these de Y. Doh : 'Bioacoustique robuste embarquee' (USTV).

Enseignements / Teaching :

Création de nouveaux cours :
Traitement automatique de la parole et applications Télécom, Transcription de documents numériques, Recherche d'information, Biométrie, RI dans le documents audio-visuels, theorie de l'information, theorie des graphes, notamment dans le cadre du master Sciences de l'Information et des Systemes (SIS) et du nouveau master Developpement et Applications sur Plateformes Mobiles MASTER DAPM.

Short bio

    H. Glotin is a professor at the Insitut Universitaire de France (since 2011) and Univ. of Toulon (since 2010), in the Systems & Information Sciences CNRS lab. He is leading the DYNI team on stochastic multimodal information retrieval. He received his master 1 in computer science from University Pierre et Marie Curie-Paris. In his master 2 thesis in artificial intelligence in Grenoble he proposed the first modelisation of vocalic system evolution, addressing the emergence of a common phonetic code in a society of communicating speech agents using evolutionary learning, which has been extended in many other works. He carried out his PhD at the Inst. of Perceptual Artificial Intelligence (IDIAP) CH and Inst. of Spoken Communication - Perception Team Grenoble on "Robust adaptive multi-stream automatic speech recognition using voicing and localization cues". In 2000 he has been involved as expert at Johns Hopkins CSLP lab with IBM human language team in audiovisual Large Vocabulary Speech Recognition. After two years as research engineer at CNRS lab on phonology and Semantic analyses, he has been an assistant professor position at University of Toulon in 2003. He then conducted researches on multimodal pattern analysis and retrieval systems, audiovisual indexing, cognitive models and bioacoustics. He is co-author of one hundred of international refereed articles, and of an international (US, CANADA...) patent on a real-time bio-acoustic indexing algorithm. He is chairing since eight years the summer school in Multimodal Information Retrieval ERMITES. His research interests include signal processing, scene Understanding (vision, audition, bioacoustics), cognitive Systems and machine learning.
  • Qualifié pour les fonctions de McF en sections 27 (Computer Science) et 61 (signal processing).
  • Ingénieur de recherche / research ingeneer CNRS @ Equipe de Recherche en Syntaxe et Sémantique - Toulouse, 2001-2002.
  • Selected for the NATO Advanced Study Institute on the Dynamics of Speech Production and Perception OTAN & ISCA grants, summer 2002.
  • Vice director of the Sciences Research and Formation Unit (UFRST) of USTV 2011-13
  • Elected at the concil of UFRST 2010-13
  • President of the research commission of the UFRST 2010-13
  • Elected at the LSIS lab. concil 2010-14
  • CRCT 50% in 2006, délégation CNRS 50% in 2010-2011
  • Leader of the team project DYNI since 2008
  • Professeur des universités depuis 2010 (HDR 2007, qualification 2008)
  • PES 2009-12 (Rank A by nat. comm.) ; PEDR 2005-08
  • Director of the computer science departement at USTV 2003-2006.
  • Member of the program committee of : Journées Etudes Parole 2002-04-06-08; RFIA 2006; CORIA 2005-06; Journées de Rochebrune de 2000 à 2005, BDA 2004-05; Majestic 2004... ; Reviewer for international journals: Speech Communication, Information Sciences...
  • Member of the computer science selection committee USTV 2004-2009, member of Pr. computer science selecting commission USTV 2011 and 2012.
  • President of InCognito science association 1996-1997
  • Co-fondateur de l'Association Francophone de la Communication Parlée (AFCP) et membre élu de son Comité Administratif (www.afcp-parole.org) 2000-2002, 2002-2004,2004-2006 (trésorier).


  • Apprendre à s'amuser : Clara & Nolan...
    Last update : April 2012

    * What is IUF ? (extracted from IUF web site) : The mission of the Institut universitaire de France is to promote the development of an high quality university research, and to strengthen interdisciplinary projects... The junior and senior members are selected by two separated international juries, since they must be composed by at least 40 per cent foreign scientists. So, it ensures a constant renewal and no one among the laureates can be co-opted...The evaluations of the juries are based on the scientific work (innovation, interdisciplinary projects, international reputation, prizes and awards, and the ability of the applicant in leading a team, in developing a project and the quality of this project)...IUF member status represents in total two per cent of the number of teaching-searchers working in the French universities.