We wanted to build something to lessen pressure of societal interactions on those who medically can't help it. Alzheimer's patients, well represented in popular media, seemed an ideal choice.

What it does

It uses a facial recognition API called Kairos to compare the face of the person standing in front with the faces of friends and family of the Alzheimer's patient, thereby letting them know with a quick brief who they're speaking to, so that they can carry on with the conversation uninterrupted with a quick memory refresh.

How we built it

We used Raspberry Pi, it's camera module and an online API system to compare the faces of the people standing directly in front of the camera at eye level.

Challenges we ran into

Installing Ubuntu on the Raspberry pi, working with and modifying the Kairos API, translating javascript code so it could work with Chromium

Accomplishments that we're proud of

Getting the facial recognition to work, installing Ubuntu on the Raspberry Pi 3

What we learned

Raspberry Pi 3, Javascript, CSS, Kairos API, HTML

What's next for FacialZheimer

Automated command systems using voice output features to read out the name of the person, more accurate facial recognition using IBM Watson API

Share this project: