Inspiration

A lot of students have trouble focusing in class as well as understanding material, and in large class settings, it is difficult for teachers and lecturers to determine weak points among their students. We aim to provide a helpful tool to aid teachers in focusing on better teaching engagement strategies.

What it does

Users (teachers/lecturers) can create sessions, which consist of an image of their audience during the lecture. Our product uses Hume AI in order to analyze these images and provide insights on how the class is feeling overall (what percentage is confused, distracted, etc). This technology can be generalized beyond the academic space and can be used as well for concerts and public speaking events.

How we built it

We created a DynamoDB database to store session information. On the frontend webpage, you can create a new session, which will be added to the database. For each session, you can upload an image. Upon uploading, the app will run Hume AI's API on it and output summarized results and insights for the user to view.

Challenges we ran into

We had some trouble accessing the database, as well as figuring out how to separate different faces in the image. With the database specifically, we had trouble with ensuring that elements from the database are deleted and show up as deleted on the frontend.

Accomplishments that we're proud of

Our product is able to differentiate faces and provide analytics for each face. We are also able to upload images for this pipeline.

What we learned

We learned how to interact with DynamoDB, as well as connect items to frontend components. We also learned how to use face recognition software in Python and how to use Hume AI's API.

What's next for EnGauge

We would like to create a much more user-friendly dashboard, as well as display more analytics for the user. We would also like to add functionality to add multiple images at once so that the user may view changes in student emotions over time.