Evaluation of Cognitive Assistance for Visual Impairment to Recognize Dynamic Walking Environments Using Vibrotactile Stimulation

This paper proposes an approach to assist the visually impaired in walking independently by presenting people flow and sidewalk edge information. To achieve this, a robust estimation of the direction of people flow, and a method for detecting the edges of sidewalks are presented. Image learning for sidewalk estimation is performed using binary semantic segmentation that classifies sidewalks and other areas to see sidewalks regardless of pedestrians and shadows on sidewalks. The effectiveness of the methods was evaluated. Then, preliminary user experiments will investigate the effect of the walking assistance information presented by this approach on the decision of walking behavior.