Digital Library

cab1

 
Title:      MAGVI: TOWARDS SALIENCY-DRIVEN VIDEO MAGNIFICATION APPLICATION FOR THE PEOPLE WITH LOW VISION
Author(s):      Prajakta Thakur, Tanay Dalvi, Varun John and Swati Chandna
ISBN:      978-989-8704-31-3
Editors:      Katherine Blashki
Year:      2021
Edition:      Single
Keywords:      Low Vision, Intelligent User Interface, Video Magnification, Video Saliency
Type:      Full
First Page:      36
Last Page:      43
Language:      English
Cover:      cover          
Full Contents:      click to dowload Download
Paper Abstract:      Interacting with dynamic content such as videos, movies is very challenging for people with low vision. While screen magnifier software enables visually impaired people to pan and zoom the static content, such as text and images. They still lack efficient and intuitive means to pan and zoom over the dynamic content such as videos, movies, or YouTube videos. Therefore, a new approach is needed to enable people with low vision to interact with the dynamic content with ease. This paper presents a concept of the fully integrated application called as MagVi which aims towards making dynamic content accessible. This is composed of two main components; the first component is the video magnification application that uses the video saliency models to automatically identify the salient Region of Interests (ROI) in different videos. The second component of this application allows visually impaired users to view the identified ROI in a mobile application and also to dynamically pan and zoom, change the font size of the text throughout the application. The quantitative and qualitative evaluation with five low vision people revealed that the application was considered effective and efficient to use. Our approach has enabled people with low vision to pan and zoom over the dynamic content from a mobile phone which indicates the potential of making videos accessible to visually impaired people.
   

Social Media Links

Search

Login