Project Description

Bionic vision is a rapidly advancing field aimed at developing visual neuroprostheses (“bionic eyes”) to restore useful vision to people who are blind. However, a major outstanding challenge is predicting which electrodes in the device should be activated to produce a desired visual image (“percept”) in the mind of the patient.

In this project, students will approach this as a regression problem, where the goal is to identify the electrode activation pattern that produces the smallest error between predicted and target image. Under the mentorship of Prof. Beyeler, and in collaboration with other members of the Bionic Vision Lab, students will develop deep learning models that an encode an image as a simulated electric stimulus and use it in combination with a brain-inspired computational model of bionic vision in an end-to-end learning task. Students will have a chance to learn about bionic vision, computational modeling, and how to design a deep learning model.

Team Members

  • Sriya Aluru
  • Gita Supramaniam
  • Harshita Gangaswami
  • Shivani Sista
  • Vanessa Salgado

Professor and Mentors

  • Prof. Michael Beyeler
  • Jacob Granley

Meeting Time

  • Meeting with Prof. Beyelor and Jacob
    • Wednesdays 3:30PM - 4PM
  • ERSP meeting with central mentors (zoom for the first two weeks, in person after that)
    • Chinmay: Wednesdays 6:30PM - 7PM
    • Diba: TBD

Links to Proposals and Presentation

  • Proposal (first draft): link
  • Proposal (after peer review): link
  • Final proposal (after instuctor feedback): link
  • Final presentation: link

Individual Logs

Peer Review

Project Documentation and Resource