VizLens

Demonstration of VizLens app
2016

Research areas

Accessibility Crowd Sourcing

VizLens: A Robust and Interactive Screen Reader for Interfaces in the Real World

The world is full of physical interfaces that are inaccessible to blind people, from microwaves and information kiosks to thermostats and checkout terminals. Blind people cannot in­dependently use such devices without at least first learning their layout, and usually only after labeling them with sighted assistance.

VizLens is an accessible mobile ap­plication and supporting backend that can help blind people use nearly any interface they en­counter by providing accurate and real-time feedback.

VizLens users capture a photo of an inaccessible interface and send it to multiple crowd workers, who work in parallel to quickly label and describe elements of the inter­face to make subsequent computer vision easier. This technology solves a long-standing challenge in accessibility by deeply integrat­ing crowdsourcing and computer vision, and foreshadows a future of increasingly powerful interactive applications that would be currently impossible with either alone. 
 

Researchers

Students

Anhong Guo
Xiang Anthony Chen