Research
See also the detailed curriculum (PDF, January 2024).
Eye Tracking
Detecting where the user is looking at is extremely useful in many situations. We apply eye tracking to create gaze-based interfaces, to improve e-learning systems, to implement gaze-based soft biometrics, to study solutions for effective information presentation, to semi-automatically evaluate the usability of user interfaces, and to study eye behavior in different contexts.
Vision-Based Perceptive Interfaces
Perceptive interfaces, able to acquire information about users and the environment in which they operate, can be very useful for human-computer interaction. In our research, we focus on exploiting computer vision to implement interfaces based on gesture recognition, head tracking, and general scene understanding.
Visual Systems for Browsing Large Collections of Images
Within the field of Information Visualization, the subclass of Information Presentation has received relatively limited attention. We are especially interested in browsing large collections of images in effective and fast ways, starting from the assumption that it is often necessary to examine image databases to search for something, but without knowing what exactly in advance.
Visual Languages
Exploiting the two- or three-dimensional space, visual languages represent an interesting alternative to textual languages in many contexts. Our work is mainly aimed at implementing effective languages for non-expert users, to accomplish tasks such as programming, action plan specification, and database querying.
E-Learning
E-learning can provide many benefits. However, to be really useful, an e-learning system should rely on proper user models and interaction paradigms, possibly also being able to identify when the user is having difficulty understanding a certain topic.
Graphical User Interfaces, Usability, and Accessibility
Almost always, using a computer means interacting with visual formalisms. However, some visual systems may lead to interaction models which are hard to control. Our efforts are focused on the development of new interaction paradigms for visual interfaces, as well as on the identification of usability and accesibility criteria for their design.
Computer Vision
The aim of computer vision is to process images acquired with cameras to produce proper representations of objects in the world. Our works are targeted at developing methods for both basic feature extraction and high-level exploitation of visual information.