department of informatics

Research of the DIVA Group

The DIVA research group (Document, Image and Voice Analysis) belongs to the Department of Informatics of the University of Fribourg and is active in several research fields dealing with multimedia engineering and machine intelligence. Our research projects are listed below, organized by research themes:

Ongoing Projects

Media Analysis and Recognition

  • Document Analysis and Recognition
  • Image Processing
    • FPID: Fribourg Product Image Database
  • Document Image Layout Ground-truthing
    • DIVADIA: Historical Document Layout Ground-Truthing framework
  • Document Representation
    • Visualization of the physical and logical structures of electronic documents

Human Computer Interaction

  • Multimodal Human-Machine Interaction
    • iMuSciCA is a pioneering approach using music for fostering creativity and deeper learning. The DIVA group introduces pen- and gesture-based interaction for music co-creation and sound analytics.
    • TEMPO : Tangible EMbodied Personal Objects
  • Gesture Analysis and Recognition
    • ARAMIS: Augmented Reality Avatar as Model for Interaction Science
    • ASGARD : Advanced System for Gesture and Activity Recognition and Detection
    • MAGI : Mind Augmented Gesture Interaction
    • FEOGARM : A Framework to Evaluate and Optimize Gesture Acquisition and Recognition Methods
    • EcoGes : Economic free hands gesture recognition
  • Emotion Recognition
  • Information Visualization and Multimedia Mining
    • Risk Manager Dashboard: visualization for assessing and managing risk in corporate networks through metrics and user behavior analysis
    • Operators for visual scaling of visualizations

Finished Projects

Media Analysis and Recognition

  • Document Analysis and Recognition
    • HisDoc 2.0: Historical Document Analysis
    • HisDoc: Historical Document Analysis, Recognition, and Retrieval
    • PAWR: Printed Multi-Font and Multi-Size Arabic Word Recognition at Ultra-Low Resolution
    • XED : A Tool for Extracting Hidden Structures from Electronic Documents
    • Dolores : Logical Structure Extraction From Electronic Documents
    • PART: Printed Arabic Recognition Text in low resolution
    • VLR-OCR : Very Low Resolution Character Recognition
  • Speech Analysis and Processing
    • FASLAV : Towards Fully Automated Spoken Language Acquisition, Understanding and Speaker Verification by Machines
  • Biometry
    • BioSecure : Biometrics for Secure Authentication
    • CHASM : Combined Handwriting And Speech Modalities
  • Biometry

Human Computer Interaction

  • IM2-HMI :Human Machine Interaction
  • Multimodal Human-Machine Interaction
    • HephaisTK : A Toolkit Allowing Rapid Creation of Multimodal Interfaces
    • MeModules : Tangible shortcuts to digital information for dailylife activities
  • Information Visualization and Multimedia Mining
    • Humanitics: Visual Analytics for All
    • WotanEye : Personal Information Management and Visualization
    • NetSecurity : Network Security Intelligence Through Distributed User Behaviour Modelling and Interactive Visualizations
    • FaericWorld : A Framework for Managing and Exploring Multimedia Archives through Links and Static Documents
    • JFriDoc : A document-centric multimedia meeting browser
    • SMAC : Smart Multimedia Archives for Conferences
  • Multimedia management and analysis
    • IM2-DMA : Database Management and meeting analysis

Document Engineering