Contact

Research

Keywords

Image Analysis Image interpretation Pattern Recognition Segmentation Information/Knowledge Perception/Computer Vision Feature Selection
Multi-agent systems Content-based image retrieval Agrégation multi-critères
Smart Markers Biomedical Imaging Document Image Analysis

Collaborations

•     L3i, University of La Rochelle
•     LIRIS(UMR CNRS 5205), INSA-Lyon / University of Claude Bernard Lyon 1 / University of Lumière Lyon 2 / École Centrale de Lyon
•     CReSTIC (EA 3804), University of Reims
•     Imageries du Vivant (IDV), Sorbonne Paris Cité(SPC)
•     Institut de recherche et d'histoire des textes (IRHT), Cnrs
•     Quantitative Imaging Laboratory, Stanford School of Medicine, Stanford University

                             

Projects

I am involved (and I have been involved) in different research projects related to biomedical imaging and document image analysis.

On going projects

HighVision Project (2025-2028) Involvement : Member Funding: ANR (France) Computer vision for a history of early news photo agencies at scale. The High-Vision project aims to advance historical investigations and research in computer vision and artificial intelligence (CVAI) for large collections of historical photographs. It seeks to harness state-of-the-art models for AI training data with a view to shape little-curated photographic archives into digital assets so that they can be investigated by historians, archivists, and the public on a large scale and in innovative ways. The objective of this project is to go beyond state-of-the-art developments in computer vision, focusing on layout segmentation as pre-processing steps, object recognition, image similarity and retrieval followed by metadata enrichment to improve the multimodal indexing steps of images from different data corpora. This High-Vision project will establish a robust pipeline that can effectively trace the circulation of news photographs, document their changing editorial contexts and accurately shape a mass of digitised historical photos without pre-existing metadata into a usable collection. Collaboration: LARCA/LIPADE/Université Paris Cité, LIP6/Sorbonne Université (SU), Maison des Sciences de l’Homme Mondes (UAR 3225), Service Historique de la Défense (SHD), Image Centre (IMC) at Toronto Metropolitan University (TMU).

Projet iaPacs (2021-2025) Involvement : PI Funding: Idf - Philips The iaPacs project focuses on the integration of a content-based image search engine in a medical image database to assist radiologists in the interpretation of images in clinical routine and decision making.

Past projects

Projet eycon (2021-2024) Involvement : Resp. partie images Funding: chaire - Idex The central objective of EyCon is to explore computational solutions to enhance the visualization, navigation and analysis of large visual corpora.

ANR TIMES (2017-2021) Involvement : Member (16 %) Funding: French Agence Nationale de le Recherche The TIMES project aims to support the analysis of satellite image time series in order to monitor the dynamics of our Planet, its land-cover and land-use. The objective of this project is to provide new high-performance processing techniques for mapping and monitoring environmental changes from massive, heterogeneous and high frequency data times series. Our team covers the image processing part of the project, with the main mission of developing and validating novel image analysis methods dedicated to environmental monitoring of landscape objects (e.g. coastal erosion, surface deformation, land cover changes) from the exploitation of the mass of available heterogeneous geospatial data (aerial and satellite images, point cloud data) acquired with very-high temporal frequency. The ANR TIMES project is a collaboration between recognized teams in Computer Sciences (LIPADE, MIPS, MONASH (Australia) and SERTIT-ICUBE), possessing expertise in image analysis, pattern recognition, machine learning and high performance computing, and in environmental geography (land cover/use change analysis) with LIVE and ICUBE-SERTIT.

ANR SHADES (2014-2018) Involvement : Member (22 %) Funding: French Agence Nationale de le Recherche The SHADES project is an interdisciplinary project on the security of documents combining experts from computer science and law. The objective of this project is to provide a new tool for authenticating a document through an advanced electronic signature in order to fight against fraud and falsification. Addressing the drawbacks of current solutions for hybrid document security, this signature, based on the document’s content and structure, will allow recognizing and authenticating a document independent from its physical representation as long as its content and/or structure have not been modified. Thanks to a hashing of the document’s information during the signature computation, no information from the original document will be deduced from its signature alone. This will be completed by a study on the legal position of this technology and its interest for legal professionals. The project will be conducted in collaboration with a trusted third party and an end user.

Programme Interdisciplinaire Imageries du Vivant: entre progrès et libertés (2014-2018) Involvement : Member (22 %) Funding: USPC - France The IDV (Imageries du Vivant) program is a program focused : on running an interdisciplinary research network bringing together colleagues and teams working on the generation / acquisition of biomedical images and on their exploitation; on the creation of an environment dedicated to the exploitation of large masses of acquired images (Big Data) for the development of new biomarkers in bioimaging; on a reflection on the ethical and legal problems raised by the exploitation of images of the living. As part of this project, we are working in particular on the construction of an intelligent atlas related to the tdomain f Interactive Content-Based Image (more generally Case) Retrieval. This work is in collaboration with the Imaging Quantitative Laboratory (Stanford School of Medicine) from Stanford University.

ANR ORIFLAMMS (2013-2016) Involvement : Member (18 %) Funding: French Agence Nationale de le Recherche The Oriflamms project (Ontology Research, Image Features, Letter form Analysis on Multilingual Medieval Scripts) The Oriflamms project brings together laboratories in the field of human and social sciences (SHS), laboratories in the field of Science and Technology of Information and Communication (STIC) and a company (A2iA) to improve the transcription of ancient documents with an interdisciplinary and innovative approach by introducing knowledge related to Medieval systems and styles of writings, according to their production context and their language. The work of the SIP team of the Lipade concerns the study of the variability of the forms contained in the writings.

OSEO DOD (2013-2016) Involvement : Associated Partner Funding: OSEO - France DOD poject (Offer of Document Dematerialization) is in partnership with the industrial company Itesoft-Yooz. The objective of this project is to identify and investigate the degradations that a document image may contain in relation to the original document. In an acquisition context (by camera or scanner) delocalized in relation to the document operator, it is therefore necessary to measure the quality of the image and to improve it as far as possible.

PIA - PiXL (2012-2014) Involvement : Member (22 %) Funding: French Investissements d'Avenir As part of the objective of creating a technology park for the dematerialisation and the promotion of digital content, this project focused on the documents produced by the BNF maps and plans department and the administrative documents provided by IteSoft. Interests are color, handwritten/printed texts, security and mobility. The task of the Lipade SIP team consisted of the handwritten/printed text separation.

ANR Graphem (2008-2011) Involvement : Member (15 %) Funding: French Agence Nationale de le Recherche The GRAPHEM (Grapheme based Retrieval and Analysis for PaleograpHic Expertise of Middle Age manuscripts) project is a multi-disciplinary project that is based on rich medieval manuscripts corpus. It has two ambitions: to contribute to the creation of a true objective and scientific paleography; to create accurate methods for accessing to the contents of these manuscripts, using word-image similarities (Word Spotting, Word Retrieval). The richness and diversity of the writings (of Latin language) will allow us to elaborate and test shape descriptors that will be used in the two applications. A particular caution will be held on the study of graphemes as pieces of strokes holding relevant information. The laboratories composing the consortium are the LIRIS, the IRHT, the LIFO, the CRIP5 and the Ecole des Chartes. They worked together a few years ago on a project aiming to the exploration of medieval manuscripts. This project was part of the "Société de l'information" program of the CRNS and was entitled "Formes et Couleurs, outils de recherche". It was in this context that the roots of GRAPHEM were constituted.

ANR NAVIDOMASS (2007-2010) Involvement : Member (25 %) Funding: French Agence Nationale de le Recherche The objective of this project is to help a humanities user to index and search large quantities of historical documents by analyzing the different elements of images and text, as well as the structure of the document. NAVIDOMASS is based on four main principles: Anywhere (global access), Anyone (public and multi-language), Any time and Any media (different modes of information transmission: Web portal, Smartphone interface, reprints of documents etc .).

Publications

Link to a ranked list of Computer Science conferences.

My page on scopus and my page on DBLP:

Journal referee

General journals

• IEEE Transactions on Image Processing
• Pattern Recognition (Elsevier)
• Pattern Recognition Letters (Elsevier)
• Fuzzy Sets and Systems

Biomedical image analysis journals

• IEEE Journal of Biomedical and Health Informatics (J-BHI)

Teaching

A. Current Course Responsibilities

Université Paris Cité (France)

@ UFR de Mathématiques et d'Informatique

• Bachelor of Computer Science- BCS (2nd year)
  • MLL3U2O : Computer Programming - C Language


• Master of Computer Sciences (First Year), Master 1
  • MMC1E31 : Fundamentals in Signal and Image Processing (co-responsibility)
  
  
  • MMF3E12 : Algorithmic Geometry
  
  
• Master of Computer Sciences, Second Year, major Vision et Machine Intelligente
  • [2018-2024] UE Project(co-responsability)
  • MMF3E72: Project Building
  • MMF3E72: Bibliography Report
  • MME3U4: Development & Implementation
  
  

@ Faculté des Sciences Fondamentales et Biomédicales

• Master of Biomedical Engineering
  • BMH109 : Computer Programming - Java & Python Languages
  • BMH303 : Medical Image Analysis I (co-responsability)
  
  

B. Administrative responsability

Université Paris Cité (France)

@ Faculté des Sciences Fondamentales et Biomédicales

• [Since 2010] Co-chair of the Biomedical Engineering Master, major BioImaging (BIM)
  The BIM track offers high-level interdisciplinary education and training supported by the complementary skills of ParisDescartes and ParisTech. A large network of research laboratories provides students access to industrial and experimental imaging systems that utilize innovative technologies.
  

@ UFR de Mathématiques et d'Informatique

• [2007-2010] Chair of the Master of Computer Science applied to Life Sciences

C. Thematic Fields taught in my career

• Image Analysis
  • Image Processing
  • Biomedical Imaging
  • Algorithmic Geometry
  • Pattern Recognition

• Computer Programming Activity
  • languages : Java, Python, C++, C, HTML, TurboPascal
  • Algorithms and Data Structures
  • Software Engineering


• Office (Word, Excel, PowerPoint)

Students supervision

I have (had) the pleasure of working with the following people:

Post-Docs

• Zografoula Vagena (12 months)
  Subject: Semantic coherence integration for optimizing imaging retrieval systems in radiology
  University: Université Paris Cité
  Funding: diip Strategic Project
  Supervisors: F. Cloppet, C. Kurtz
  Period: January-December 2023
  


• Van Cuong Kieu (18 months)
  Subject: Blur Estimation and correction in document images
  University: Université Paris Cité
  Funding: Itesoft Company
  Supervisors: Cloppet, N. Vincent
  Period: 2015- 2017
  


• Sameh Hamrouni (9 months)
  Sujet: Handwritten/Printed text separation
  University: Université Paris Descartes
  Funding: Pixl - Investissement d'Avenir
  Supervisors: F. Cloppet, N. Vincent
  Period: 2014
  

Ph.D. students - Doctoral program

• Agathe Roullet (PhD in progress)
  Sujet: Multimodal deep learning model for the segmentation of cancerous lesions from clinical reports
  University: Université Paris Cité
  Funding: Dassault Systèmes
  Supervisors: F. Cloppet, C. Kurtz
  Period: 2025 - 2028
  


• Xioyang Wei (PhD in progress)
  Sujet: Combining visual and textual information for enhancing pathologic case retrieval systems in radiological practices
  University: Université Paris Cité
  Funding: China Scholarship Council (CSC)
  Supervisors: F. Cloppet, C. Kurtz
  Period: 2022 - 2025
  


• Thibault Sauron
  Sujet:Search for similar pathological cases in radiological image databases
  University: Université Paris Cité
  Funding: IDF-Philips
  Supervisors: F. Cloppet, C. Kurtz
  Period: 2021- 2024
  


• Héloïse Alhéritière
  Subject: Sécurisation des documents hybrides par une analyse physique des contenus
  University: Université Paris Descartes
  Funding: ANR SHADES
  Supervisors: N. Vincent (co-doctoral advisor), J.M. Ogier (co-doctoral advisor), F. Cloppet (30%), C. Kurtz
  Period: 2015 - 2018
  


• Soumaya Trabelsi
  Subject: Analyse et suivi des anomalies mammaires à partir d'imagerie multimodales IRM/mammographie
  University: Université Paris Descartes (France) - ENIS (Tunisie)
  Funding: Grant from Ministère de l'Enseignement Supérieur et de la Recherche Tunisien
  Supervisors: L. Wendling (co-doctoral advisor), D. Sellami Masmoudi (co-doctoral advisor), F. Cloppet (25%)
  Period: 2014 - 2017
  


• Arnaud Boucher
  Subject: Recalage et analyse d'un couple d'images : Application aux mammographies
  University: Université Paris Descartes (France)
  Funding: Cifre - Fenics
  Supervisors: N. Vincent(Doctoral Advisor), F. Cloppet (50%)
  Period: 2009 - 2013
  


• Hassan Chouaib
  Subject: Sélection de caractéristiques: méthodes et applications
  University: Université Paris Descartes (France)
  Funding: ANR Navidomass
  Supervisors: N. Vincent (Doctoral Advisor), F. Cloppet (50%)
  Period: 2008 - 2011
  


• Sylvain Berlemont
  Subject: Analyse d'image d'ADN peigné, extraction des entités linéaires et extraction de patterns
  University: Université Paris 5 (France)
  Funding: Cifre - Genomics Vision
  Supervisors: N. Vincent (co-doctoral advisor), J.C. Olivo-Marin (co-doctoral advisor - Institut Pasteur), F. Cloppet (20%)
  Period: 2005 - 2008
  


• Elodie Dusch
  Subject: Détection, localisation et suivi de spots modélisés par mixture de gaussienne
  University: Université Paris 5 (France)
  Funding: Institut Pasteur Corée
  Supervisors: N. Vincent (co-doctoral advisor), A. Genovesio (co-doctoral advisor - Institut Pasteur Corée), F. Cloppet (10%)
  Period: 2005 - 2008
  


• Estelle Glory
  Subject: Segmentation de cellules observées par microscopie en contraste de phase
  University: Université Paris 5 (France)
  Funding: Cifre - Celogos
  Supervisors: G. Stamon (co-doctoral advisor), J.C. Olivo-Marin (co-doctoral advisor - Institut Pasteur), F. Cloppet (30%)
  Period: 2002 - 2005
  


• Santiago Venegas-Martinez
  Subject: Analyse et segmentation de séquences d'images en vue d'une reconnaissance de formes efficace
  University: Université Paris 5 (France)
  Funding: Grant from Mexican Governement
  Supervisors: G. Stamon (Doctoral advisor), F. Cloppet (20% - last year)
  Period: 1999 - 2002
  


• Emmanuel Augustin
  Subject: Reconnaissance de mots manuscrits par un système hybride : Modèles de Markov Cachés et Réseaux de Neurones
  University: Université Paris 5 (France)
  Funding: Cifre - A2iA
  Supervisors: G. Stamon (Doctoral advisor), F. Cloppet (20% - last year)
  Period: 1998 - 2001
  


• Mohammed Borchani
  Subject: Caractérisation, recherche et appariement d'images par attributs de texture
  University: Université Paris 5 (France)
  Supervisors: G. Stamon (Doctoral advisor), F. Cloppet (30% - last year)
  Period: 1997 - 2001
  


• Yann Dieulangard
  Subject: Compression sélective d'images segmentées : Application au codage d'images infrarouges pour le renseignement
  Funding: Cifre - Aerospatiale
  University: Université René Descartes (France)
  Supervisors: G. Stamon (Doctoral co-advisor), D. Barba (Doctoral co-advisor), F. Cloppet (20% - last year)
  Period: 1995 - 1998
  


• Vannary Meas-Yedid
  Subject: Analyse de cartes scannées : interprétation de la planche de vert
  Funding: MESR - Ministère de la Recherche et de l'Enseignement Supérieur (France)
  University: Université René Descartes (France)
  Supervisors: G. Stamon (Doctoral advisor), F. Cloppet (20% - last year)
  Period: 1995 - 1998
  

Master (M.Sc.) - Research internship

• Samuel Gonçalves
  Cursus: EPITA - M2 master en informatique, parcours IMA (Sorbonne Université France)
  Subject: Multimodal representation learning for identifying image circulation in large collections of historical photographs
  Funding: ANR HighVision project Encadrant-e(s): F. Cloppet, C. Kurtz, I. Bloch, D. Foliard
  Période: 2025, 6 months
  


• Fei Fei
  Cursus: Master Informatique VMI (Université Paris Cité)- France)
  Subject: Identification of image circulation by AI in large collections of historical photographs
  Funding: diip M2 project Encadrant-e(s): D. Foliard, F. Cloppet, C. Kurtz
  Période: 2024, 6 mois
  


• Zixuan Wang
  Cursus: Ecole Centrale de Nantes (France)
  Subject: Early fusion of multiple MRI sequences for enhancing pathologic case retrieval systems in radiology
  Funding: diip M2 project Encadrant-e(s): C. Kurtz, F. Cloppet
  Période: 2022, 6 mois
  


• Guillaume Sérieys
  Cursus: Master BME (Université Paris Cité, France)
  Subject: Combining visual and textual information for enhancing image retrieval systems in radiological practices
  Funding: diip M2 project Encadrant-e(s): C. Kurtz, F. Cloppet
  Période: 2021, 6 mois
  


• Alexandre Landrodie
  Cursus: Master Informatique Informatique-VMI (Université Paris Descartes, France)
  Subject: Classification d'images de textes en manuscrit ou imprimé à l'aide de réseaux de neurones
  Funding: IMDS Encadrant-e(s): F. Cloppet, N. Vincent
  Période: 2020, 6 mois
  


• Walid Amaieur
  Cursus: Master Informatique IP (University Paris Descartes, France)
  Subject: Table extraction in document images
  Supervisors: Héloïse Alhéritière, C. Kurtz, F. Cloppet, N. Vincent
  Period: 2018, 4 months
  
• Jordan Gonzales
  Cursus: Master Informatique IMA (University Pierre and Marie Curie, France)
  Subject: Content-based image retrieval of multisequence MRI with re-ranking
  Supervisors: C. Kurtz, F. Cloppet and Laure Fournier (partnership with APHP-HEGP, project Imageries du Vivant-USPC)
  Period: 2016, 6 months
  
• Lylia Yakouben
  Cursus: Master Informatique IP (University Paris Descartes, France)
  Subject: Détection de signature dans une image de document
  Supervisors: Héloïse Alhéritière, C. Kurtz, F. Cloppet, N. Vincent
  Period: 2016, 4-5 months
  


• Yanki Sesyilmaz
  Cursus: Master Informatique IMA (University Pierre and Marie Curie, France)
  Subject: Content-based image retrieval of multisequence MRI
  Supervisors: C. Kurtz, F. Cloppet and Laure Fournier (partnership with APHP-HEGP, project Imageries du Vivant-USPC)
  Period: 2016, 6 months
  


• Cao Deng Feng
  Cursus: Master Biomedical Engineering (University Paris Descartes, France)
  Subject: Fusion of EPR and Micro CT imaging to obtain full-scale anatomical and functional information
  Supervisors: F. Cloppet, Yves Frapart
  Period: 2016, 2 months
  


• Pineiro Zarate Ivan
  Formation: Master 1 Biomedical Engineering (Université Paris Descartes, France)
  Sujet: Radiological image retrieval in big databases (PACS): Integration of specific global quantitative features focusing on local regions of interest (ROI) in the images.
  Supervisors: F. Cloppet, C. Kurtz
  Period: 2016, 2 months
  


• Héloïse Alhéritière
  Cursus: Master Informatique IP (University Paris Descartes, France)
  Subject: Extraction de la structure d'un document digitalisé
  Supervisors: N. Vincent, F. Cloppet, C. Kurtz
  Period: 2015, 5 months
  


• Neila Adili
  Cursus: Master Informatique RIP-Image (University Paris Descartes, France)
  Subject: Quantifier la variabilité des écritures dans un contexte d'adaptation d'algorithmes dans un système de transcription automatique des zones d'écriture de documents du Moyen-Âge
  Supervisors: N. Vincent, F. Cloppet
  Period: 2014, 6 months
  


• Frédéric Moreau
  Cursus: Master Informatique RIP-Image (University Paris Descartes, France)
  Subject: Analyse d'images écho-cardiographiques
  Supervisors: N. Vincent, F. Cloppet
  Period: 2013, 6 months
  


• Mejdi Djerbel
  Cursus: Master Informatique RIP-Image (University Paris Descartes, France)
  Subject: Recalage d'images multimodales RPE et MicroCT
  Supervisors: F. Cloppet, Yves Frapart
  Period: 2011, 6 months
  


• Houssem Gueziri
  Cursus: Master Informatique RIP-Image (University Paris Descartes, France)
  Subject: Segmentation d'images ultrasonores
  Supervisors: F. Cloppet, N. Vincent
  Period: 2011, 6 months
  


• Arnaud Boucher
  Cursus: Master Informatique PCC-Image (University Paris 5, France)
  Subject: Étude et Élaboration d'un algorithme de segmentation d'images biologiques
  Supervisors: F. Cloppet
  Period: 2007, 5 months
  


• Isabelle Boudoux
  Cursus: DESS Informatique Appliquée aux Sciences de la Vie (University Paris 5, France)
  Subject: Utilisation de la texture pour la segmentation d'images biologiques
  Supervisors: F. Cloppet
  Period: 2001, 6 months
  


• Anne-Laure Lefeuvre
  Cursus: DESS Informatique Appliquée aux Sciences de la Vie (University Paris 5, France)
  Subject: Segmentation d'images de lymphocytes obtenues par microscopie optique
  Supervisors: F. Cloppet
  Period: 2000, 6 months
  



Engineering degree - Research internship

• Carlos Tor Díez
  Cursus: Engineering degree in Telecommunication (Universitat Politècnica de Catalunya, BarcelonaTech, Barcelona, Spain)
  Subject: Integrating relevance feedback in content-based image retrieval applications for medical imaging
  Supervisors: C. Kurtz, F. Cloppet (and partnership with Télécom ParisTech)
  Period: 2015, 5 months
  


• Manel Ben Ameur
  Cursus: Projet de Fin d'Etudes ENIS, Ecole Nationale d'Ingénieurs de Sfax (Tunisie)
  Subject: Quantifier la variabilité des écritures dans un contexte d'adaptation d'algorithmes dans un système de transcription automatique des zones d'écriture de documents du Moyen-Âge
  Supervisors: N. Vincent, F. Cloppet
  Period: 2015, 4 months
  


• Paul-André Idoux
  Cursus: Engineering degree in Computer Science (EPITA Graduate School of Computer Science, Paris, France)
  Subject: Large-scale content-based image retrieval in medical imaging
  Supervisors: C. Kurtz, F. Cloppet
  Period: 2014, 6 months
  


• Pierre-Paul Couka
  Cursus: Engineering degree in Computer Science (ESME-Sudria Graduate School of Computer Science, Paris, France)
  Subject: Segmentation d'images couleur en vue de caractériser le degré d'interaction de protéines au cours du cycle cellulaire
  Supervisors: F. Cloppet
  Period: 2006, 6 months