ReFlick

Inspiration

In our journey towards a more sustainable world, we must adhere to proven practices aimed at preserving our Earth. One such practice we have targeted for improvement is recycling. Through discussions with my peers, it has become evident that recycling is not a priority in the United States. This can be attributed to various factors, with convenience being the most significant. Whether it's a lack of awareness about recycling or limited access to recycling facilities, our objective is to leverage technology to make recycling more accessible and convenient.

What it does

Users upload or take pictures to educate themselves about the nature of the material and how to recycle it properly, as well as explore alternatives to minimize its environmental impact.

How we built it

By employing Machine Learning, an image captured by the user undergoes image classification, enabling the model to identify the object. Subsequently, AI is employed to educate the user on sustainable practices related to the identified material.

Challenges we ran into

We were very indecisive when it came to the overall structure of the project. As we worked through the project we realized other ideas might be better and easier. Because of this, the backend went through implementing many different AI models eventually settling on using OpenAI.

Accomplishments that we're proud of

We are proud of the simple and effective design of the project. In addition, the accuracy of our image classifier.

What we learned

Throughout the project, we were exposed to many API and image classification models. We also gained a deeper understanding of using React and Flask to implement a Web Application. Lastly, we gained experience in deploying a project on Vercel and AWS.