Personal tools
  • Log in
You are here: Home Demo material
Navigation
The GazeCom project is funded by the European Commission (contract no. IST-C-033816) within the Information Society Technologies (IST) priority of the 6th Framework Programme.
 
Document Actions

Demo material

by Michael Dorr last modified 2010-06-01 12:03

Gaze-contingent interactive displays

gcid.png

Gaze-contingent interactive displays (GCID) are devices that interact with the human observer to optimize visual communication systems. Such a display actively tracks the observer’s gaze position – where s/he specifically is looking - and changes its display output in order to guide the observer’s gaze to task-relevant locations. We coined the term Gaze Guidance for this type of human-computer interaction. We imagine that GCIDs can be useful in a wide variety of tasks, such as working at a PC display, analyzing medical images, interacting with virtual worlds, or even driving a car (cockpit).

Here we show some results from our research:

Eye movements on natural videos

ducks_gaze.jpg

A video of a canal scene near Lübeck, Germany. Overlaid are the gaze positions of 46 observers: in a relatively uneventful scene, observers look anywhere. However, when a swarm of ducks flies across the scene, every observers’ gaze is drawn to a single spot.

GCID with local contrast reduction

gcid_demo.jpg

On this gaze-contingent interactive display, contrast was reduced at some interesting locations to render them less distracting, that is to manipulate the observer's gaze to other, potentially interesting locations.

Gaze guidance to teach experts' skills in the natural sciences

balisti01.jpg

This video, with a voice-over by a biology professor, is used to teach biology students about different fish locomotion patterns (here: balistiform locomotion). To understand and learn the specific pattern, an observer's gaze is directed to the relevant image region by blurring the irrelevant (and potentially distracting) regions.

This experiment was performed in collaboration with the Knowledge Media Research Center, Tübingen.

Gaze-contingent Virtual Reality environment

tunnel.jpg

nvisor.jpg

To test the task-relevance of gaze guidance, subjects have to perform a challenging perceptuo-motor task in this Virtual Reality environment.

Design of Virtual Reality environment

tunnelcreation.jpg

The graphics of the Virtual Reality environment are put together in Blender, physics are simulated the NVIDIA PhysX engine, and the Object Oriented Input System is used for navigation.

 

Powered by Plone CMS, the Open Source Content Management System

This site conforms to the following standards: