Virtual Human Interface Group

Ongoing projects

Ambient Facial Interfaces (AFI)

Ambient Facial Interfaces provide visual, non-verbal feedback via photo-realistic animated faces. They display emotional facial expressions, and body language most reliably recognizable by people. These digital faces are controlled by the output parameters of physical measurements or data derived from the state of the user or the products and objects he or she is interacting with. The output of an AFI system combines these measurements into a single facial expression that is displayed to the user, thereby allowing them to evaluate overall "quality" at at-a-glance.

Ambient Assited Living (AAL)

This project aims to develop a novel life style and health management platform for elderly people living alone or with their families. It focuses on creating an open-architecture and open source service model where home-based and wearable (yet medically accurate and reliable) sensors provide data and information to a portable controller (such as a PDA or Internet Tablet) which not only collects them, but evaluate trends, provide advice on health, diet or workout regiment, and provides a transparent and redundant data link between the elderly person and his or her care takers. Besides connectivity the system also utilizes a AFI technology (see above) to deliver non-verbal feedback and encouragement in a readily discernable manner independent of language or culture.

Interactive Virtual Child

We created an interactive art installation, called "FacePlay", with a the goal of demonstrating a new kind of user interface in a playful manner for children and adults alike. At the center of this application is a virtual baby face reacts to the visitors with its elaborate repertoire of facial expressions. the VHI system was configured to use two cameras placed at different heights. The first one is at the same level as that of a child's head sitting on top of the monitor showing the life size digital baby face. The second one is located higher to be able to collect images from the accompanying adults. The system runs face recognition algorithms on both of these image streams to determine not only if there are users to interact with, but also whether it is a child alone, with an accompanying adult or just a grown up standing in front of the installation. Based on this information the VHI uses different strategies (e.g. child like behavior or more adult actions) to address the people standing in front. The facial expressions of the digital character are linked to the faces as they move in within the field of view of the cameras. This from of non-verbal feedback based on visual perception is supplemented by the tactile subsystem in the form of a touch screen located in front of the virtual face. As the child reaches for the virtual baby's face, it notices when being touched, reacts accordingly and lets the user put finger paint on by changing the color and properties of the underlying skin.

Portable Virtual Reality in Education

Our research focuses on developing Virtual reality-based education systems that employ a student-centered interaction model based on the theory of closed-loop emotional modulation (CLEM). CLEM works by mimicking the interaction process that take place between people during everyday dialogues and conversations. To implement this goal our portablr VR system comprises of several state-of-the-art key modules. Each of these modules play a vital part in the process of information exchange and supports the interactive learning experience. Our goal was to create a general architecture that can be freely and easily configured to address the various needs of a large variety of VR application in the classroom. While different solutions may rely-on various configurations of these modules they all share the same underlying system. Therefore developing new educational content is simple and very efficient.