Patrick Perez

Distinguished Scientist

Imaging Science Lab Researchers


  • CNN-News: image retrieval benchmark of key frames from news videos (with VLAD descriptors), on request
  • TSFT: face tubes from TV series (589 manually croped and annotated face tubes of 94 subjects)
  • Hannah: dense person annotation of Woody Allen's Hannah and her sisters
  • Scratch: annotated data on scractch detection for movie restoration, with TelecomParisTech
  • VideoInpainting: videos and objects' masks for object inpainting, with TelecomParisTech

Lectures (since 2010)

  • Deep learning for the movie industry - UPMC, Paris, Nov. 2017
  • Graph signals and neural nets - Inria Rennes, Setp.2017 | Inria Grenoble, Oct. 2017
  • From processing to learning on graphs - Google Brain, Montreal, June 2017 | Google Research, Zurich, April 2017 | Imaging in Paris, March 2017
  • From video objects to visual effects - Willow Inria/ENS seminar, Paris, Jan. 2017
  • On visual comparison - GCPR, Hannover, Sept. 2016 | StatLearn, Vannes, April 2016
  • Visual tracking of arbitrary objects - GDR Isis on Multi-Target Tracking, Paris, Oct. 2015
  • From image to video inpainting with patches - GDR Isis on Missing Data, Marseille, Oct. 2015 | MissData, Rennes, June 2015 |Tel-Aviv University, Jan. 2015 |ICISP'14, Cherbourg, June 2014 | JMAIB' 14, Bordeaux, April 2014
  • Manipulating videos objects: from removal to retargeting - Inria Sophia Antipolis, Sept. 2015
  • Learned visual embeddings - Allegro Workshop, Inria, Grenoble July 2015
  • Face2face: comparing faces, with applications - Oxford Robotics Seminar, Oct. 2014 | INRIA Vision micro-Workshop, Rennes,  Oct. 2014
  • Visual tracking - Vision and Sport Summer School (VS3), Pragues, Aug. 2014 | VS3, Pragues, Aug. 2012 | Inria-Grenoble, Dec. 2011 | VS3,  Zurich, Aug. 2011
  • Computer vision in everyday life - Centrale Marseille,  June 2014 | June 2013 | Dec. 2011
  • From images to descriptors, and back again - MIA'14, Paris, January 2014
  • A glimpse at visual tracking - Inria/ENS VRML summer School, Paris, July 2013
  • Dense long-range correspondences in videos - Inria-Grenoble, Oct. 2012
  • On visual search - MPI Saarbruken, Sept, 2012 |Oxford Brookes, March 2012 | Stanford, April 2011
  • Geodesic Image and Video Editing - MPI, Saarbrucken, June 2011 | ENS Cachan, May 2010
  • Viewpoint and Pose Invariant Video Analysis - International Computer Vision Summer School, Sicily, July 2010

Selection of Publications (since 2010)

(see complete list on GoogleScholar)

Conferences and workshops

Short Bio

Patrick Pérez received the engineering degree from Ecole Centrale Paris in 1990, and the Ph.D. degree from University of Rennes in 1993. After one year as a postdoc in the Dpt of Applied Mathematics at Brown University (USA), he joined Inria (France) in 1994 as a full time researcher. From March 2000 to February 2004, he was with Microsoft Research (Cambridge, UK). He then returned to Inria as a senior researcher and took, in 2007, the direction of Vista research team of the Inria Rennes Center. In November 2009, Patrick Pérez joined Technicolor where he leads exploratory research on computer vision and image analysis. He is a Technicolor Fellow. He is currently on the Editorial Board of the International Journal of Computer Vision.

Former editorial duties

  • Associate Editor for the IEEE Transactions on Image Processing (2002-2006)
  • Associate Editor for the IEEE Transactions on Pattern Analysis and Machine Intelligence (2008-2012)
  • Area chair for ECCV (2008, 2010, 2012, 2014), ICCV (2011, 2012), BMVC (2017)

Research interests: Models and tools for understanding, analysing, processing and manipulating still and moving images.

Projects: FP 7 EU Integrated Project AXES: Access to Audiovisusal Archives (Jan. 2011 - March 2015).