Inspiration
We were inspired (traumatised) when we were studying for our Y1S1 finals. Everyday we were just staring at code, or math on our computer screens with no end in sight. Naturally, our mood was very bad. Hence came the idea to build an app to lift our mood admits intense studying/working sessions! (Or, maybe, mess with you somewhat...)
What it does
Monitors users mood through facial expression monitoring using webcam, then gives appropriate prompts accordion to mood detected. For example if the user is sad it will prompt the user to cheer up.
Secret mode: Users can toggle to this mode to have a little fun! Under this mode any facial expression other than being completely neutral will lead to ___ generating certain entertaining notifications...
How we built it
We used Facial Expression Recognition model (FER), currently the most accurate model publicly available....
Challenges we ran into
Syncing the model with the javascript code was really difficult, as if they were not synced up JS would receive junk data, even though the FER model was working properly.
Accomplishments that we're proud of
We successfully got around the challenge mentioned above and implemented it into a web-based programme.
What we learned
How to fine tune the model for the most accurate and consistent results, and how to use the results in a meaningful way.
What's next for NUSlack
To make NUSlack a well rounded mental health bot, by leveraging health gadget like Apple Watch to monitor not just facial expressions, but also important bodily functions like heart rate and body temperature and give more accurate and useful feedback for our users.
Log in or sign up for Devpost to join the conversation.