Vision-Based Perceptive Interfaces

Although the WIMP interaction paradigm (based on Windows, Icons, Menus, and Pointing devices) has been very useful for providing a common face to computing, new interaction options are now possible in the direction of "naturalness". Perceptive interfaces can acquire, by means of proper sensors, information about the user and the surrounding environment. In particular, Vision-Based perceptive Interfaces (VBIs) employ video cameras to implement machine vision. In addition, or as alternatives, other devices can be exploited, such as ToF (Time of Flight) video cameras or tools similar to the Microsoft Kinect sensor (also used in biometric applications).

Marco Porta has considered the two main kinds of interaction identifiable for VBIs (depending on the active or passive role played by the user), namely explicit and implicit communication. While in explicit communication users are fully aware that their actions (e.g., gestures) will be interpreted in a certain way by the computer, in implicit communication users are indirectly "observed" to obtain information on what they are doing or what is happening in the nearby environment; or, also, to get information about their "emotional states", through the recognition of facial expressions and body movements. Of course, interfaces based on Eye Tracking are a particular case of perceptive interfaces.