Inspiration

We were inspired by the theme of Space. We wanted to make it about exploring and going out to discover things that you are not used to. That is why we used some technologies that we were not very familiar such as the CLIP model which we used to be able to do image recognition in pytorch. We were super inspired with going out of our comfort zone and were really happy with its payoffs.

What it does

With SpacePals you can embark on a journey to discover the cosmos! Take pictures of all the alien animlas that you find along the way, and look back at them later on in your Explorers's Guide! You can see your progress there as you progress on your journey and learn some cool information about all the animals you encounter. 👽

How we built it

We used the CLIP model and gave it names of things that it should be able to recognize. We then built the frontend and connected it using GraphQl. We worked on getting the ability to take images and then have our model on the backend categorize what the image is showing. It is then update in your traveler's guide to show that you have found the alien.

Challenges we ran into

We ran into multiple challenges. Ranging from being able to take a picture on our frontend, and having it work with our model in the backend. Additionally, being able to deploy the frontend and backend was really challenging. Most of our challenges were with being able to get things to get some features set up, but once they were, things were pretty smooth.

Accomplishments that we're proud of

We are really proud of the project as a whole. We are proud of the model that we were able to use (CLIP) in order to get our images to be classified. Additionally, we are happy with the interaction that it has with the frontend and the way that everything is set up. Overall we learned a lot from this experience and are really happy that we did it because we were able to learn about machine learning and just overall good engineering skills.

What we learned

We learned all about doing frontend development, and being able to work with a more intricate backend than we are used to. Additionally, we learned a lot about machine learning models and how they can be utilized in conjunction with other parts of a project. We learned about many different libraries that can be used to make a nice looking frontend and also some techniques to deploy a backend and frontend.

Share this project:

Updates