Abstract
Advances in artificial intelligence over the past decade, combined with increasingly affordable computing power, have made new approaches to accessibility possible. In this chapter we describe three ongoing projects in the Department of Computer Science at North Carolina State University. CAVIAR, a Computer-vision Assisted Vibrotactile Interface for Accessible Reaching, is a wearable system that aids people with vision impairment (PWVI) in locating, identifying, and acquiring objects within reach; a mobile phone worn on the chest processes video input and guides the user’s hand to objects via a wristband with vibrating actuators. TIKISI (Touch It, Key It, Speak It), running on a tablet, gives PWVI the ability to explore maps and other forms of graphical information. AccessGrade combines crowd-sourcing with machine learning techniques to predict the accessibility of Web pages.