Michaël Clément

Associate professor of computer science

Research
LaBRI, UMR CNRS 5800
michael.clement@labri.fr
351 cours de la Libération, 33400 Talence, France
Teaching
Bordeaux INP - ENSEIRB-MATMECA
michael.clement@enseirb-matmeca.fr
1 avenue du Dr. Albert Schweitzer, 33400 Talence, France

About me

I am associate professor of computer science (maître de conférences) at Bordeaux INP, in the ENSEIRB-MATMECA engineering school, and researcher at the LaBRI laboratory, in the Image and Sound team. I obtained my PhD in computer science from Université Paris Descartes, where I was advised by Laurent Wendling and Camille Kurtz. My work was about modeling and learning spatial relations between objects for image understanding, with applications in document analysis, remote sensing and medical imaging. After my PhD, I was a postdoctoral researcher at the Centre for Vision Research of York University, where I worked with James Elder on shape data analysis for 2D and 3D reconstruction problems.

Short CV

[Since 2018]
Associate professor of computer science, Bordeaux INP, Bordeaux, France
Research at the LaBRI, in the Image and Sound team
Teaching at the ENSEIRB-MATMECA engineering school
[2017–2018]
Postdoctoral researcher, Centre for Vision Research, York University, Toronto, Canada
Project: Sparse dictionary learning from shape data
Advisor: James Elder
[2014–2017]
PhD in computer science, LIPADE, Université Paris Descartes, Paris, France
Thesis title: Modeling and learning spatial relations for image recognition and understanding
Advisors: Laurent Wendling and Camille Kurtz
Teaching assistant at the UFR Mathématiques et Informatique
[2013–2014]
Master's degree in computer science, Université Paris Descartes, Paris, France
Research internship at the LIPADE
Advisors: Laurent Wendling and Camille Kurtz
[2010–2013]
Engineer's degree in computer science and applied mathematics, Grenoble INP - Ensimag, Grenoble, France
Software engineer apprentice, STMicroelectronics
[2008–2010]
DUT informatique (two-year undergrad in computer science), Université Paris Sud, Orsay, France
Software developer apprentice, IBM Delivery Services

Publications

Preprints (1)

AssemblyNet: A large ensemble of CNNs for 3D whole brain MRI segmentation
Pierrick Coupé, Boris Mansencal, Michaël Clément, Rémi Giraud, Baudouin Denis de Senneville, Vinh-Thong Ta, Vincent Lepetit and José V. Manjón
arXiv preprint, 2019.

Articles in international journals (4)

Multi-scale superpatch matching using dual superpixel descriptors
Rémi Giraud, Merlin Boyer and Michaël Clément
Pattern Recognition Letters, Special Issue on Hierarchical Representations, 2020 (accepted).

Fuzzy directional enlacement landscapes for the evaluation of complex spatial relations
Michaël Clément, Camille Kurtz and Laurent Wendling
Pattern Recognition, 2020 (in press).

Learning spatial relations and shapes for structural object description and scene recognition
Michaël Clément, Camille Kurtz and Laurent Wendling
Pattern Recognition, vol. 84, pp. 197-210, 2018.

Directional enlacement histograms for the description of complex spatial configurations between objects
Michaël Clément, Adrien Poulenard, Camille Kurtz and Laurent Wendling
IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 12, pp. 2366-2389, 2017.

Articles in international conferences (6)

Enlacement and interlacement shape descriptors
Michaël Clément, Camille Kurtz and Laurent Wendling
(accepted) International Conference on Pattern Recognition and Artificial Intelligence (ICPRAI), 2020 (Zhongshan, China).

AssemblyNet: A novel deep decision-making process for whole brain MRI segmentation
Pierrick Coupé, Boris Mansencal, Michaël Clément, Rémi Giraud, Baudouin Denis de Senneville, Vinh-Thong Ta, Vincent Lepetit and José V. Manjón
International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI), 2019 (Shenzhen, China).

Local enlacement histograms for historical drop caps style recognition
Michaël Clément, Mickaël Coustaty, Camille Kurtz and Laurent Wendling
IAPR International Conference on Document Analysis and Recognition (ICDAR), pp. 299-304, 2017 (Kyoto, Japan).

Fuzzy directional enlacement landscapes
Michaël Clément, Camille Kurtz and Laurent Wendling
International Conference on Discrete Geometry for Computer Imagery (DGCI), pp. 171-182, 2017 (Vienna, Austria).

Bags of spatial relations and shapes features for structural object description
Michaël Clément, Camille Kurtz and Laurent Wendling
International Conference on Pattern Recognition (ICPR), pp. 1994-1999, 2016 (Cancún, Mexico).

Color object recognition based on spatial relations between image layers
Michaël Clément, Mickaël Garnier, Camille Kurtz and Laurent Wendling
International Conference on Computer Vision Theory and Applications (VISAPP), vol. 1, pp. 427-434, 2015 (Berlin, Germany).

Abstracts in international conferences (2)

Human and machine perception of three-dimensional shape from contour
James H. Elder, Yiming Qian, Michaël Clément and Srikumar Ramalingam
European Conference on Visual Perception (ECVP), 2019 (Leuven, Belgium).

What are the sparse components of 2D shapes?
Michaël Clément and James H. Elder
European Conference on Visual Perception (ECVP), 2018 (Trieste, Italy).

Articles in national journals (1)

Description d’objets en couleurs à partir des relations spatiales entre régions structurelles
Michaël Clément, Camille Kurtz and Laurent Wendling
Revue des Nouvelles Technologies de l'Information. Fouille de Données Complexes, RNTI-E-31, pp. 57-72, 2016.

Articles in national conferences or workshops (3)

Sacs de relations spatiales et de formes pour la reconnaissance d'images de scènes naturelles
Michaël Clément, Camille Kurtz and Laurent Wendling
Journées francophones des jeunes chercheurs en vision par ordinateur (ORASIS), 2017 (Colleville-sur-Mer, France).

Descripteurs directionnels d'enlacement et d'entrelacement entre objets
Michaël Clément, Camille Kurtz and Laurent Wendling
Congrès National sur la Reconnaissance des Formes et l'Intelligence Artificielle (RFIA), 2016 (Clermont-Ferrand, France).

Descripteurs de relations spatiales entre régions structurelles pour la reconnaissance d'objets en couleurs
Michaël Clément, Camille Kurtz and Laurent Wendling
Atelier Fouille de Données Complexes (AFDC), Conférence Internationale sur l'Extraction et la Gestion des Connaissances (EGC), pp. 25-36, 2015 (Luxembourg, Luxembourg).

PhD thesis

Modélisation et apprentissage de relations spatiales pour la reconnaissance et l'interprétation d'images
Michaël Clément
Université Paris Descartes, LIPADE (EA 2517), defended on September 26, 2017.

Teaching

Since 2018, I teach computer science at ENSEIRB-MATMECA, a public engineering school located in Bordeaux. I am teaching courses mainly related to:

  • C programming
  • Object oriented programming
  • Software engineering projects
  • Image processing, computer vision
  • Artificial intelligence, machine learning
See the syllabus of the computer science department. I am also head of first year for the RSI work-study programme.

From 2014 to 2017, I was a teaching assistant at UFR Mathématiques et Informatique of Université Paris Descartes. I have been involved in the following classes (in French):