Project Info

WasteWisely

This project did not submit a demo video on Devpost.

Inspiration

The way we throw trash has a lot of impact on our environment. Not recycling recyclables or throwing landfill trash into recycling bins all cause negative impacts on our planet. However, now many people know how to classify trash correctly. This kind of knowledge is even harder for children to obtain. Thus, we developed WasteWisely, an application that helps people classify any trash that they have conveniently.

What it does

When one scans an image of any object using our app, our AI engines will tell you whether the object is (most likely) recyclable, disposable, or compostable. The app also gives engaging explanations about how to dispose the object that you scanned.

How we built it

We use ReactNative to develop UX/UI and Flask for the backend. We also use Google Cloud Vision AI to detect an object from the camera's picture. After getting what kind of object the image has, we use ChatGPT API to suggest how to handle the peace of waste best.

Challenges we ran into

Our projects have multiple components: ReactNative, Flask, Google Cloud Vision, ChatGPT API. We develop each aspect on a different computer. Thus, we ran into a lot of errors when trying to combine all components for deployment. For example, we got errors when sending images from the front-end to the back-end, or we got authentication errors when running ChatGPT API on a different computer.

Accomplishments we're proud of

We were proud that we have successfully managed all the complexities of full-stack development and AI technologies to develop a meaningful application for the planet. Our team also has two members who participated in a hackathon for the first time, but we were able to split our work equally and we all enjoy the process of developing the app.

What we learned

We learned how to incorporate Google Cloud Vision AI and ChatGPT APIs to our application to harness the power of these technologies. We also learned a lot about mobile development and full-stack development in general.

What's next

We still need to make the UX/UI look more engaging to attract young users. Also, the accuracy of pre-trained AI models is somewhat low for our project's objectives. Thus, we want to look into ways to customize our AI models to achieve higher accuracy.

Analysis

Compare with all teams

View

Metric

Figures cover GitHub contributors during the hackathon window. A co-authored commit counts in full for each author, so per-member totals add up to more than the whole-team figures.

Technology

Found in codeClaimed only

ExpressIn code
FlaskIn code
JavaScriptIn code
LangChainIn code
OpenAIIn code
PythonIn code
PyTorchIn code
ReactIn code
TensorFlowIn code

9 of 9 appear in the indexed code.

AI coding agents

No AI coding agent signals were found in this repository.

Detected from committed agent config files and commit authorship. Absence of a signal is not proof an agent was unused.

Codebase size

Source size

12 KB

Source files

Counts recognized source files only; vendored directories, binaries and lockfiles are excluded, so this is smaller than the repository on disk.