Getting in Touch: Virtual Maps for the Blind
From: Scientific American - 04/2007
By: Rachel Ross

Researchers at the Aristotle University of Thessaloniki in Greece have
developed a way to transform video into virtual haptic maps that can provide
the blind with a better grasp of cities and building layouts.
Three-dimensional models are sometimes used as maps for the blind, but can
only accommodate one person at a time, and paper maps with ridges cannot
provide comprehensive information. However, with the Greek system digital
dioramas are available to people around the word and are accompanied by
audio. After shooting video of an architectural model, each frame of the
video is then processed using software designed by lead researcher
Knostantinos Moustakas and his team. As the camera angle changes throughout
the video, each structure is analyzed by the software to determined its shape
and orientation. The result is a 3D grid of force fields representing each
structure. "Each point on the grid has an associated force field," says
Moustakas. The human interface consists of the CyberGrasp glove, which pulls
on the user's fingers, and the Phantom Desktop, a wand that applies pressure
to the user's hand in one direction. Moustakas has also designed a system
that converts paper maps into 3D maps; users would run their finger or a wand
down grooves representing streets as the names are said aloud to them. In
tests of 19 subjects, the dioramas were found to be preferable for small
groups of structures, and the 3D street maps were preferable for large areas. 

Read the entire article at:
http://www.sciam.com/article.cfm?articleID=BDC29B20-E7F2-99DF-3C6EC4B5008F1FE9&chanID=sa001

Links:
Extraction of 3D Scene Structure from a Video for the Generation of 3D Visual
   and Haptic Representations
http://www.w3.org/2004/02/mmi-workshop/moustakas-crth.pdf

CyberGrasp
http://www.immersion.com/3d/products/cyber_grasp.php

Phantom Desktop
http://www.sensable.com/haptic-phantom-desktop.htm