Inspiration

Beagles are one of the most vocal animals. BeagleBoard follows the same principles of being able to vocalize things you want to experience with your friends.

What it does

It's a collaborative vision board that allows the user to turn their text to image and then add that image to the board. For example, you have always wanted to go on a yacht in monaco. You would say "yacht in monaco" on the prompt screen, which would then add the most relevant yacht image to the board. Also, say your friend has always wanted to go skydiving. He would add "skydiving" to the prompt and a relevant sky diving image would be added to the board.

How we built it

It's built using React with firebase backend and unsplash api. For each prompt, it essentially searches up the most relevant image for that search but submitting a request through unsplash api. Then, a link to that image is stored in the firebase database, which updates the vision board in real-time without having to use a large amount of storage. Bootstrap css functions are used for a good chunk of the UI, making it both web/mobile accessible.

Challenges we ran into

One challenge was that the image would take a long time to load after the user had submitted the prompt. This was happening because I was saving the image to firebase storage and then displaying it every time. To solve this, I started just saving a public link to the image to a database and just calling that link every time the page was loaded, making the app much faster.

What's next for BeagleBoard

One thing I want to tackle is being able to see who added a particular image on the board. I do like the anonymity of the current app where you can't tell which one of your friends added the particular image, but this would be avenue I would like to explore. Another thing would be to generate an image instead of searching for one using generative AI models. This was one of the first thing I looked into but the problem with this approach is just the sheer amount of processing time, making it unfeasible. However, if we bring the processing time down, it would be perfect for BeagleBoard.

Built With

Share this project:

Updates