Claire-Hélène Demarty - Senior Scientist - Technicolor Rennes

Claire-Hélène Demarty

Senior Scientist

Location:Rennes

Laboratory:Imaging Science Lab

What's new:

The benchmark campaign MediaEval 2014  is starting. This year again, Technicolor  will participate to the organisation of the Affect Task on Violent Scene Detection, as advisor for the task.

The Technicolor Violent Scenes Detection (VSD) Dataset is freely available for the research community.

Short Bio:

I graduated from Telecom Paris Tech in 1994 and received a Ph.D. degree in Computer Science, Mathematical Morphology from Mines Paris Tech in 2000. I joined the Technicolor Research & Innovation Center in 2004 as a senior researcher and became senior scientist in 2010. Located in Rennes, France, I am working on multimedia indexing technologies. Prior to Technicolor, I worked at LTUTechnologies (2000 – 2002) and at INRIA Rennes – IRISA (2003 – 2004), a French public research center, as a researcher in image analysis and video indexing technologies. I am author or co-author of more than 15 papers and I am holding several patents.

Research interests:

  • Image processing (video indexing, image segmentation),
  • Computer vision,
  • Multimodal analysis (multimodal video structuring, multimodal event detection),
  • Machine Learning and classification,
  •  Mathematical morphology.

Projects:

Currently working at the Technicolor R&I Rennes center, I am a member of the Image Editing and Enhancement Technical Area.

My research is currently focusing on new ways of summarizing and representing multimodal content.

In the past, i was in charge of the activity of Violent Scenes Detection in movies, which is still participated in the MediaEval framework.

MediaEval is a benchmarking initiative dedicated to evaluating new algorithms for multimedia access and retrieval. MediaEval emphasizes the multimodal character of the data. In MediaEval, Technicolor was co-organizing the Affect Task on Violent Scenes Detection between 2011 and 2013.

Quaero was a program promoting research and industrial innovation on technologies for automatic analysis and classification of multimedia and multilingual documents, in which Technicolor was strongly involved. I was personally contributing to the following Quaero projects: Core Technologies Cluster, Corpus, PVAA++ project.

 

Latest publications

Article

VSD2014: A Dataset for Violent Scenes Detection in Hollywood Movies and Web Videos

Markus Schedl, Mats Sjöberg, Ionut Mironica, Bogdan Ionescu, Vu Lam Quang, Yu-Gang Jiang, Claire-Hélène Demarty

CBMI2015
June 15, 2015

Benchmarking Violent Scenes Detection in Movies

Claire-Hélène Demarty, Bogdan Ionescu, Yu-Gang Jiang, Vu Lam Quang, Markus Schedl,

CBMI 2014
June 18, 2014

Variability Modelling for Audio Events Detection in movies

, Claire-Hélène Demarty, Guillaume Gravier, Patrick Gros

Journal of Multimedia Tools and Applications - Special Issue on Content Based Multimedia Indexing
May 01, 2014

Multimodal Violence Detection in Hollywood Movies: State-of-the-Art and Benchmarking

Claire-Hélène Demarty, , Bogdan Ionescu, Guillaume Gravier, Mohammad Soleymani

Fusion in Computer Vision
April 01, 2014

VSD, a public dataset for the detection of violent scenes in movies: design, annotation, analysis and evaluation

Claire-Hélène Demarty, , Mohammad Soleymani, Guillaume Gravier

MTAP
January 01, 2014

Technicolor/INRIA team at the MediaEval 2013 Violent Scene Detection Task

, Claire-Hélène Demarty, Guillaume Gravier, Patrick Gros

MediaEval 2013
November 01, 2013

The MediaEval 2013 Affect Task: Violent Scenes Detection

Claire-Hélène Demarty, , Markus Schedl, Bogdan Ionescu, Vu Lam Quang, Yu-Gang Jiang

MediaEval 2013 Workshop
October 18, 2013

Audio Event Detection using Multiple Audio Words and Contextual Network

, Claire-Hélène Demarty, Guillaume Gravier, Patrick Gros

CBMI2013
June 17, 2013

A benchmarking campaign for detecting violent scenes in movies

Claire-Hélène Demarty, , Mohammad Soleymani, Guillaume Gravier

ECCV : Workshop on Information fusion in Computer Vision for Concept Recognition
October 29, 2012

Electrodermal activity applied to violent scenes impact measurement and user profiling

Technicolor/INRIA/Imperial College London at the MediaEval 2012 Violent Scene Detection Task

, Claire-Hélène Demarty, Guillaume Gravier, Patrick Gros

MediaEval 2012 Workshop
October 04, 2012

The MediaEval 2012 Affect Task: Violent Scenes Detection in Hollywood Movies

Claire-Hélène Demarty, , Guillaume Gravier, Mohammad Soleymani

MediaEval 2012 Workshop
October 04, 2012

De la détection d’évènements sonores violents par SVM dans les films

, Claire-Hélène Demarty, Guillaume Gravier, Patrick Gros

orasis2011
June 09, 2011

Paper

The MediaEval 2014 Affect Task: Violent Scenes Detection

Mats Sjöberg, Bogdan Ionescu, Yu-Gang Jiang, Vu Lam Quang, Markus Scheld, Claire-Hélène Demarty

MediaEval 2014 workshop
October 18, 2014

Electrodermal activity applied to violent scenes impact measurement and user profiling

IEEE SMC 2012
October 14, 2012

Multimodal information fusion and temporal integration for violence detection in movies

, Claire-Hélène Demarty, Guillaume Gravier, Patrick Gros

ICASSP 2012
March 25, 2012

The MediaEval 2011 Affect Task: Violent Scenes Detection in Hollywood Movies

Claire-Hélène Demarty, , Guillaume Gravier, Mohammad Soleymani

MediaEval 2011 Workshop
September 01, 2011

Learning temporal modality integration with Bayesian Networks

, Claire-Hélène Demarty, Guillaume Gravier, Patrick Gros

MediaEval 2011 Workshop
September 01, 2011