Inspiration

Our team, consisting of bilinguals and trilinguals, recognizes that to easily pick up a new language, daily exposure and interaction are crucial. However, not everyone has the opportunity to immerse themselves in a foreign country. Hence, we created LANGlass to provide users with a means of learning a new language without the hassle of buying multiple books and online classes.

What it does

LANGlass enables quick and easy object classification and identification by simply looking at the world around you. By incorporating Ad Hawk’s eye-tracking hardware and our own visual mapping into our build, we can accurately track the user's gaze and display it on a screen, enabling identification using machine learning. Further, by labelling and pronouncing identified objects in a user-set language, we introduce users to the opportunity to learn a new language in their daily lives.

How we built it

We used Ad Hawk's eye-tracking hardware and developed software (including some cool camera calibration) to be able to map the eye's gaze in the video stream provided by a mounted camera. From there, we used machine learning to classify and identify objects around the user.

Challenges we ran into

One of the main challenges we encountered while building LANGlass was the very limited amount of time we had to create the product. During the start of our hackathon, we faced many setbacks due to hardware issues and pivoting, leading us to work on five different ideas. However, in the end, we made a choice to focus on creating something we would be proud of, regardless of how far we progressed in the hackathon. This led us to settle on making LANGlass.

Accomplishments that we're proud of

We're proud of how we conquered all the failures we faced along the way with all sorts of different projects, and still managed to pull through with a fully working product we are proud of having made.

What we learned

Through HTN, we learned that Software Development can be very difficult if team members don't share the same final vision in mind. Going forward, I believe the most crucial part of a team is the ability to communicate effectively and work together well beforehand.

What's next for LANGlass

We believe the next steps for LANGlass could include AR interfaces and more accurate eye-tracking and mapping algorithms.

Built With

Share this project:

Updates