RGB-D computer vision techniques for simulated prosthetic vision

Abstract

Recent research on visual prosthesis demonstrates the possibility of providing visual perception to people with certain blindness. Bypassing the damaged part of the visual path, electrical stimulation provokes spot percepts known as phosphenes. Due to physiological and technological limitations the information received by patients has very low resolution and reduced dynamic range. In this context, the inclusion of new computer vision techniques to improve the semantic content in this information channel is an active and open key topic. In this paper, we present a system for Simulated Prosthetic Vision based on a head-mounted display with an RGB-D camera, and two tools, one focused on human interaction and the other oriented to navigation, exploring different proposals of phosphenic representations.