Raphaël Marée

Projects

2005-today: "Bioinformatics and Modeling"

Since 2005, I am the GIGA Bioinformatics platform manager (scientific head: Prof. L. Wehenkel). In close collaboration with the Bioinformatics and Modeling research unit, we offer services to academic as well as to industrial researchers including classification of biological/biomedical data obtained from various medical instrumentation (SELDI mass spectra, microarrays, clinical databases, images, ...) based on machine learning methods. More info.

In this context, my personal focus is on the development and application of machine learning and computer vision methods for large-scale, biomedical image datasets, and on the delivery of user-friendly softwares to help biologists/pathologists/etc. to explore their imaging data and support their diagnosis.

               

2002-today: "Automatic Image Retrieval, Classification, and Semantic Segmentation"

This research is about the development and evaluation of generic methods for automatic image retrieval, classification, and semantic segmentation.

In 2003, we proposed to combine ensemble of randomized decision trees with random extraction of subwindows (square patches) described by their raw pixel values for image classification/categorization. Since then it was evaluated on a large variety of problems. It was first described in our JDS03 paper (in french) and then in our AI-2003 paper. The main contributions are summarized in the CVPR05 paper.
See our other publications for details, various applications (in particular on biomedical problems), and latest extensions for content-based image retrieval (ACCV07, IPSJ-CVA09) and semantic segmentation (BENELEARN07, VISAPP09).
The java software PiXiT implements the CVPR05 method for image classification and I suggest to use it with its default parameters as a baseline method on new datasets. If results are satisfactory, then you do not need to develop a new method. ;-)

Keywords:

• Machine learning, image classification, categorization, pixel-based, appearance-based, classifier, local, frames, random, windows, patches, subwindows, ensemble of decision trees, extra-trees, MNIST, COIL-100, AR, ORL, OUTEX, ZuBuD, TSG-20, WANG, ETH-80.

Examples of image classification problems I have been working on with the same approach:

>

Publications

To browse other publications from our group, please go here.

To access my publications on the institutional repository (ORBI), please go here.

Software

PiXiT, a Java software which implements the aforementioned image classification method is available upon request for evaluation and non-commercial purpose, in collaboration with PEPITe.

Conference Calendar

A browsable conference calendar (Machine Learning, Computer Vision, Bioinformatics) or the .ics iCalendar file.
Tip: click on too see the calendar by month.

Please note that it is not an official calendar so always refer to official conference websites to check updates and precise time schedules.

Étudiants

Propositions de sujets de TFE pour l'année académique 2009-2010 (new)

Divers

Je suis aussi passionné par la musique (sélections radio, organisation d'événements: Panoptica 2006 - 2005 - 2004 - 2003 - 2002).

La photo de temps en temps:

(Corée, ACCV 2004) (San Diego, CVPR 2005)

(Pékin, ICCV 2005) (Tokyo, ACCV 2007)

(Pierre Geurts is not only the father of Extra-Trees, he also made Yanis)