Getting fast, free, and anonymous answers to questions asked by people with visual impairments

Abstract

Much of the information we encounter in daily life is visual. For people with visual impairments, this information is not inherently accessible. Technology can be used to provide access to this visual information, through automated tools like currency identifiers, object recognizers, and optical character recognition, which allow for immediate access to visual information in a users' environment. However, these tools are often are limited to a specific domain, and can fail in non-ideal conditions (e.g., an optical character recognition tool trying to identify handwritten text). By incorporating people into the information access pipeline, human-powered tools can allow for more complex information to be extracted from the user's surroundings. Human-powered access tools connect people who have disabilities to other people who can access information on their behalf. While people may not be as quick as automated tools, the person answering can make inferences based on prior knowledge and experiences, ask for clarification, and reason over the information provided. In this article, we explore the long-term public deployment and lessons learned from VizWiz Social, a human-powered access tool that connects people with visual impairments to sighted workers or friends and family members who can answer their visual questions. By analyzing the use of this service, and running controlled experiments to determine how users value different answer sources, we have been able to build upon the original VizWiz design to create social microvolunteering, a method for getting fast, free, and anonymous answers to visual questions.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 6 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.