Hervé Bredin received his PhD on talking-face biometric authentication from Telecom ParisTech (France) in 2007.
In 2008, he joined Dublin City University (Ireland) to work on video summarization techniques applied to user-generated content.
He has been a permanent CNRS researcher since October 2008 -- at IRIT (Toulouse, France) in the SAMoVA group until September 2010, and at LIMSI (Orsay, France) in the Spoken Language Processing group since then.
His current research interest concerns multimodal fusion approaches, applied to several tasks such as semantic concept detection in videos, (multimodal) speaker identification in multimedia documents and the extraction of the narrative structure of fictional audiovisual documents (such as TV series and movies).

News

  • 2015-01-26 | Available internships
  • 2013-04-05 | Call for papers - SLAM 2013
  • 2012-10-13 | Two new papers dealing with multimodal fusion
  • 2012-06-11 | Unsupervised Speaker Identification using Overlaid Texts in TV Broadcast
  • 2012-06-11 | New release: PyAnnote 0.2.2
  • 2012-01-04 | Segmentation of TV Shows into Scenes using Speaker Diarization and Speech Recognition
  • 2012-01-04 | Community-driven Hierarchical Fusion of Numerous Classifiers Application to Video Semantic Indexing
  • 2011-10-17 | Introducing PyAFE toolkit
  • 2011-02-10 | Visualizing SGE Job Dependency Graph
  • 2008-12-19 | Automatic editing of personal movies
  • 2008-09-23 | Video footprints
  • Google Scholar profile
    LinkedIn profile
    e-m@il
    lastname@limsi.fr
    mail
    LIMSI-CNRS
    BP 133
    F-91403 ORSAY CEDEX
    phone
    +33 (0) 1 69 85 81 84