Inspiration
After the opening ceremony, our team noticed that many attendees were unaware of how to sort their trash correctly. We saw people mixing different types of waste into the wrong bins, highlighting a significant gap in understanding. This sparked our inspiration to create RetroCycle—a project designed to educate users about proper waste classification through an engaging, arcade-themed experience.
What it does
RetroCycle was built using Python for the back-end machine learning model and JavaScript/HTML/CSS for the front end. We curated a diverse dataset of trash images for training the model and designed an interactive interface that captures the fun of retro arcade games.
How we built it
We developed RetroCycle using Python for the back-end machine learning model and JavaScript/HTML/CSS for the front end. We used a dataset of trash images to train our model and designed a user-friendly interface that captures the retro arcade aesthetic.
Challenges we ran into
One major challenge was assembling a comprehensive dataset to ensure accurate classification. Integrating the model with the user interface also required careful troubleshooting to ensure a seamless user experience. Through these challenges, we learned the importance of collaboration and adaptability.
Accomplishments that we're proud of
We successfully created a functional prototype that accurately classifies trash and provides users with clear guidance on recycling. We also implemented a fun, engaging design that encourages users to learn more about waste management.
What we learned
In developing RetroCycle, we gained valuable skills in machine learning and image recognition. We learned how to train a model to classify various types of trash accurately, while also enhancing our front-end development abilities to create a user-friendly interface.
What's next for RetroCycle
Moving forward, we plan to expand the dataset to include more waste categories and improve the model’s accuracy. We also aim to incorporate user-generated content, such as a feature that allows users to submit new trash images for classification.
Log in or sign up for Devpost to join the conversation.