Raphaël Marée

me    PhD in computer science (since 2005)
   Cytomine scientific coordinator

   E-mail: Raphael.Maree@ulg.ac.be
   Tél: (+32) 4 366 26 44
       Systems and Modeling (Montefiore Institute & GIGA Research)
       Montefiore Institute (B28)
       Quartier Polytech 1
       10, Allée de la découverte
       4000 Liège

Curriculum vitae - Projects & Research Interests - Publications - Software - Conference Calendar - Students - Misc

Projects & Research Interests

(Bio)Image informatics

My personal research interests lie within the field of image informatics (in particular "bioimage informatics"), ie. the development and application of machine learning, computer vision, and software development methodologies to ease the exploitation of large images, e.g. the recognition and quantification of cells, tissues and other biological or biomedical "objects" in large-scale bioimaging datasets (e.g. in digital pathology), and on the delivery of user-friendly softwares to help scientists (biologists/pathologists/computer scientists/etc.) to exploit big imaging data, derive new knowledge, and support their findings and diagnoses.

Since October 2010, I am the initiator and scientific coordinator of the CYTOMINE project which lead to the development of the CYTOMINE open-source software platform, a "Google Maps"-like rich internet application for remote visualization, annotation and automated analysis of high-resolution, multi-gigapixels images.

See Cytomine page for related publications, and below for an illustration of some applications.

cytomine applications

Machine learning and computer vision

This research is about the design of generic methods for automatic image classification, retrieval, and semantic segmentation. Indeed, as potential applications of image recognition technologies are multidinous, we seek to develop general-purpose methods for the recognition of various types of images that share some visual regularities, without relying on too strong assumptions about patterns to recognize and acquisition conditions, and without having to rely on domain experts to design specific features.

This book chapter (2013) summarizes our work where we combine ensemble of randomized decision trees with random extraction of subwindows (square patches) described by their raw pixel values.

More recently, we have conducted large-scale empirical studies on many classification datasets to better identify influential design choices and draw general guidelines for future use, and to further increase robustness of the approach. On several datasets, results are significantly improved compared to our previous works, see our paper in Pattern Recognition Letters (2016).


Between January 2005 and October 2014, I was the GIGA Bioinformatics platform manager (scientific head: Prof. L. Wehenkel). In close collaboration with the Bioinformatics and Modeling research unit, this platform offers software development and data analysis services to academic (within the GIGA research center and beyond) as well as to industrial researchers. Its services include classification of biological/biomedical data (SELDI mass spectra, microarrays, clinical databases, ...) obtained from various medical instrumentation based on machine learning methods.
This activity lead to co-authorship of several journal papers (in Journal of Immunology, Proteomics, Annals of the rheumatic diseases, ...) in various application domains (inflammatory diseases, cancer, ...).

Open science

I'm very much in favor of basic principles of open science (open access, open data, open source, open hardware) although I still have much to learn about it.


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

Our citations according to Google Scholar are here.

Most of my publications are directly available, others (with publisher's constraints, symbolized by the padlock) are also available after filling a simple request form:


CYTOMINE is a "Google Maps"-like rich internet application for remote visualization, collaborative and semantic annotation and automated analysis of high-resolution images. It is released under an open-source software license. It also includes a data mining module in Python with our latest algorithm developments for image classification, semantic segmentation, and landmark detection, and a Java module for content-based retrieval.

PiXiT, a Java software which implements our CVPR 2005 image classification method is available upon request for evaluation and non-commercial purposes, in collaboration with PEPITe. Newer extensions and improvments are not included in this version.

Conference Calendar

For years, for personal usage I'm trying to maintain a unofficial conference calendar (.ics file) in the fields of computer vision, bioimage informatics, machine learning, biomedical imaging.

In the field of computer vision, the conference listing from USC is far more complete.


Please contact me for Master final thesis proposals, traineeships, ... in bioimage informatics or machine learning for computer vision. Especially if you are interested to be involved in open-source software development using machine learning and modern web technologies and their applications to real-world (big) image data.


I listen very much to music (mostly ambient/electronica/modern classical/fields recordings) like in this selection.
I used to select and play music on a local radio and I co-organized the Panoptica festival with friends.
I also enjoy reading (society essays and "independent" comics), eating (from thaï/indian/lebanese/vietnamese/japanese food to belgian boulet-frites), traveling, taking pictures, thinking, ...

Seoul(Korea) San Diego(San Diego) San Diego(Beijing) Tokyo(Tokyo-Kyoto) Lisbon (Lisbon) Belle-Île(Brittany)

Université de Liège Faculté des Sciences Appliquées GIGA