
Seeings humans
Beyond the confines of keyboard and mouse, new modalities for human-computer
interaction such as facial expressions, voice, gestures, and force
feedback are emerging. A very important aspect previously ignored is the
emotional aspect. The goal of our research is to automatically detect and
perceive the
affective state of a user to allow for a more natural human-computer
interaction. We apply novel computer vision and machine learning
techniques to recognize emotions from visual and audio information,
and from a joint audio-visual space.
The website "Glad or Sad" shows some results of this research.
Tracking humans
We aim to track humans in real-life outdoor circumstances robust to all kinds of weather, from one surveillance camera to another.
Participants
|