INTUIDOC : Intuitive user interaction for documents

The Intuidoc team focuses its research work on handwriting, gesture (2D and 3D) and documents under various aspects: analysis, recognition, composition and interpretation. We are also interested in human-document interaction and pen-based/gestural man-document interaction. This research relates to the handwriting and the documents under different forms: digital manuscript or printed paper form, graph, images, heterogeneous documents, etc.

The roadmap of the IntuiDoc team is on the frontier of several research axes: Pattern recognition, Machine-Learning, Artificial Intelligence, Human-Machine Interaction, Uses and Digital Learning. The aim is to explore new scientific challenges of the domain of the Human-Document Interaction with a specific focus on interactive, incremental and evolving learning, by integrating the user in the decision loop of all the processes of analysis and interpretation.

Today, four major emerging scientific axes are investigated with strong partnerships with national and international laboratories and companies:

• On-line evolving cross-learning of 2D (touch and pen {based) and 3D gestures (Kinect and Leap Motion);
• On-line analysis of drawing, sketching and handwriting with pen-based tablet for digital learning (e-education);
• Interactive learning of document structure without ground-truth;
• Document collection analysis for big-data.