Inspiration
We were inspired by relatives who had the need for a clothing organization tool.
What it does
The app allows users to log in, submit photos for processing, and view their past submitted clothing.
How we built it
We built the frontend with React and Material UI for the styling. We built the backend with AWS API Gateway, Lambda functions, S3 buckets for image storage, and Amazon RDBS for our database. Auth0 is used for user authentication, allowing them to access a persistant closet storage system
Challenges we ran into
We ran into challenges integrating API calls to OpenAI and Gemini models. It was often difficult to test the key functionality of our web app due to rate limiting.
Accomplishments that we're proud of
We are proud of the level of detail that the models are able to extract and generate from images. We believe that the art style of the generated images, simplicity of the interaction features, and overall aesthetic of the web app leads to a very relaxing and helpful user experience. We are also proud of our Auth0 integration.
What we learned
We learned a lot about utilizing the Gemini API for image and text generation. Additionally, we strengthened our skills in building multi-page React apps.
What's next for Track
A next step for the project would be to integrate additional APIs into the web app, for the purpose of sourcing related clothing items that are for sale or that others are looking to buy.
Log in or sign up for Devpost to join the conversation.