The purpose of the project is to develop a prototype for a wearable device that can detect users’ emotions. The implementation of the device will be composed of three major parts: collecting electromyography (EMG) signals from three sensors, storing and analyzing the collected data utilizing machine learning, and displaying the prediction on a smartphone application.
https://www.youtube.com/watch?v=Cc9PWFPVIpw&feature=youtu.be
We were able to implement a prototype of the device that can predict three types of emotions: anger, happy and sad with up to 70% of accuracy. More details are included in the Documentation folder of the repo.
Overview |
Mobile Application's UI |