Communications of the ACM
A New Data Trove Could Teach Computers to Tell Blind People What They Need to Know
University of Texas at Austin researchers have published a database of images and challenged the machine-vision community to use it to train machines to aid those with visual disabilities.
Credit: Technology Review
Researchers at the University of Texas at Austin (UT Austin) are publishing a database of 31,000 images along with questions and answers about them, and challenging the machine-vision community to use this dataset to train machines as effective assistants for those with visual disabilities.
The dataset comes from the VizWiz application designed by Carnegie Mellon University scientists to help the blind.
The UT Austin team analyzed photos collected by VizWiz, and then presented the images and questions to Amazon's Mechanical Turk workers to supply a short-sentence answer.
A preliminary analysis of the data offers unique insights into the challenges machine vision faces in providing this kind of assistance.
"We introduce this dataset to encourage a larger community to develop more generalized algorithms that can assist blind people," say the researchers. "Improving algorithms on VizWiz can simultaneously educate more people about the technological needs of blind people while providing an exciting new opportunity for researchers to develop assistive technologies that eliminate accessibility barriers."
From Technology Review
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA
No entries found