Abstract
Presentation Description :
Current retinal prostheses offer a promising solution to people with degenerative diseases of the outer retina. Notwithstanding, the obtained visual acuity is still severely limited and offers very basic vision restoration. In this lecture I will describe the main limitations of current retinal prosthetic approaches and will describe the concept of "active sensing", where super-resolved vision is achieved by actively scanning the visual scene. To simulate prosthetic vision reading, we displayed pixelated word images at different phosphene densities and contrasts. We measured word recognition rate and reading speed for various phosphene densities and contrasts with and without active sensing or zoom. Our results show that active scanning can significantly increase reading performance in simulated prosthetic vision and yield super-resolving capabilities. These results highlight the importance of a patient interactive interface and shed light on techniques that can greatly enhance prosthetic vision quality.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.