Low Vision Support through Eye Tracking Analysis

Abstract

Public signs are set up for the convenience of people. Often, pictograms are used instead of letters to catch the attention (for guidance, notification, and other purposes). Using eye-tracking technique, we have experimentally examined the ease of finding pictograms by low vision people. We got a conclusion that they hardly notice these signs. The purpose of our project is to conduct a visibility study wherein we wish to analyze low vision through eye-tracking analysis. We show a preliminary result of pictogram recognition in the view by using pattern matching, linking to a future system for guiding the gaze to a missing pictogram and zooming into it.