Barcelona Media. Centre d'Innovació.

Alexandre Hervieu

Current position

Since Sept. 2009: Post-doctoral position, Barcelona Media, University Pompeu Fabra, Barcelona, Spain, Stereoscopic Inpainting.
Supervisor: Vicent Caselles

 

Educational background

2005 - 2009 : Ph.D thesis (INRIA grant) with the VISTA group at the INRIA - Rennes Centre Bretagne Atlantique, Rennes, France. "Trajectory-based video event detection".
Supervisors Patrick Bouthemy and Jean-Pierre Le Cadre
http://www.irisa.fr/vista/Equipe/People/Alexandre.Hervieu.html

2004 - 2005 : Research Master of Science in Fundamental and Applied Mathematics, university of Rouen, France. Thesis on level-set analysis, creation of a toolbox creation for image processing.
Supervisor: Christian Gout

2004 - 2005 : Engineering degree in Applied Mathematics and Computer Sciences from the INSA (National Institute of Applied Sciences), Rouen, France.

 

Research

-STEREOSCOPIC INPAINTING
Issues related to image and video inpainting using depth information. Several views of a same scene are used to extract depth information, further used to process filling of missing regions.

-TRAJECTORY-BASED VIDEO UNDERSTANDING

Issues related to dynamic video content understanding and video surveillance: clustering and recognition of events and human gestures, detection of unexpected events, activity temporal segmentation, using single or several mobile object trajectories (2D or 3D); and object shape recognition and segmentation.

-STATISTICAL MODELING

Issues related to markovian models architecture, parameter estimation for reduced sets of data.

 

Collaborations

- since Sept. 2009 :
Participation to the European research project 2020 3D Media.

- 2005 - 2008 :
Participation to the European research project MUSCLE.

- 2005 - 2009 :
Participation to the European research project VISIONTRAIN.

 

Publications

 

PhD Thesis

A. Hervieu. Analyse de trajectoire vidéos à l'aide de modélisations markoviennes pour la reconnaissance de contenus . Université de Rennes I - INRIA Rennes-Bretagne Atlantique, March 2009.

 

Book Chapters

A. Hervieu and P. Bouthemy. Understanding Sports Video Using Players Trajectories. To be published  in Springer-Verlag series "Studies in Computational Intelligence", special issue on "Intelligent Video Event Analysis and Understanding".

 

Journals

A. Hervieu, P. Bouthemy and J-P. Le Cadre. A statistical video content recognition method using invariant features on object trajectories. IEEE Transations on Circuit and Systems for Video Technology, vol. 18, n°11, November 2008, "Special Issue on Event Analysis in Videos".

A. Hervieu, P. Bouthemy and J-P. Le Cadre. Reconnaissance d'événements vidéos par l'analyse de trajectoires à l'aide de modèles de Markov . Traitement du Signal, vol.26, nº 3, may 2009.

 

International conferences

A. Hervieu, N. Papadakis, A. Bugeau, P. Gargallo and V. Caselles. Stereoscopic image inpainting: distinct depth maps and images inpainting. International Conference on Pattern Recognition, ICPR'10, Istanbul, Turkey, august, 2010.

A. Hervieu, P. Bouthemy and J-P. Le Cadre. Trajectory-based handball video understanding. Conference on Image and Video Retrieval, CIVR'09, Santorini, Greece, july, 2009.

A. Hervieu, P. Bouthemy and J-P. Le Cadre. Activity-based temporal segmentation for videos of interacting objects using invariant trajectory features. Proceedings of the IEEE International Conference on Image Processing, ICIP'08, San Diego, US, october, 2008.

A. Hervieu, P. Bouthemy and J-P. Le Cadre. Video event classification and detection using 2D trajectories. Proceedings of the International Conference on Computer Vision Theory and Applications, VISAPP'08, Funchal, Portugal, january, 2008.

A. Hervieu, P. Bouthemy and J-P. Le Cadre. A HMM-based method for recognizing dynamic video contents from trajectories. Proceedings of the IEEE International Conference on Image Processing, ICIP'07, San Antonio, US, september, 2007.

National conferences

A. Hervieu, P. Bouthemy and J-P. Le Cadre. Reconnaissance d'événements dans des vidéos par l'analyse de trajectoires à l'aide de modèles de Markov . GRETSI, Troyes, September, 2007.