Environmental Sensing Technologies for Visual Impairment - Introduction to the Special Thematic Session
Year: 2016
Type: article
Abstract: Environmental Sensing Technologies for Visual Impairment Introduction to the Special Thematic Session Roberto Manduchi 1 and James Coughlan 2 Baskin School of Engineering, University of California, Santa Cruz [email protected] Smith-Kettlewell Eye Research Institute, San Francisco, CA [email protected] Peoples ability to move around, interact with objects, space and people, and engage in purposeful activities hinges in large part on their visual sense. Tasks such as orienting oneself in a hallway, finding an item at the market, or recog- nizing a friend from a distance may become difficult or impossible for those who cannot see, or cannot see well. Simple assistive devices may play an important role in supporting activities of daily living for people with visual impairment. For example, people who are blind use simple tools such as a long cane or a dog guide to move safely along a desired path, or a screen reader to access a computer. Persons with low vision often rely on magnifying glasses, telescopes, and screen enlargers to access textual information. In recent years, the widespread diffusion of smartphones, equipped with ac- cessible interfaces, has spurred a number of innovative technical solutions to some of the problems faced by visually impaired persons. Some of the most suc- cessful applications rely on remote sighted helpers for visual interpretation. For example, TapTapSee allows a blind user to take a picture and receive a textual description of the scene content, such as the main visible objects. BeMyEyes uses a FaceTime-like audio/video connection to enable blind users to share the video stream taken with their iPhone or iPadwith a remote sighted volunteer, who can answer specific queries, describe what is visible in the video, and also advise the blind user about where to move the camera for a better view. Other models are currently being explored; for example, Aira provides a remote vi- sual interpretation service by professional helpers, who are specifically trained to support blind users in tasks such as navigation, information access, and even people recognition. These types of crowdsourcing mechanisms have been quite successful for spe- cific tasks. However, crowdsourcing may not be the solution for all types of visual interpretation tasks. For once, video communication requires a good data con- nection, which is not always available, and can quickly drain the battery of a smartphone. In addition, the time to establish a connection or to receive an an- swer to a query may be substantial, discouraging frequent use of this technology. Automatic visual analysis by means of computer vision algorithms (executed on the smartphone itself, or on a remote cloud server) is an attractive alternative. Many blind people, for example, use currency reader smartphone apps such as LookTel. Optical character recognition (OCR), a mature technology originally
Authors: Roberto Manduchi, James M. Coughlan
Language: English
Subfield: Occupational Therapy
Field: Health Professions
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W2767366504