If we’re being honest, we really don’t know the climate impacts the objects we see daily cause. For example, what do we know about the CO2 footprint of bottled water? Well… based on the most cited research it's about 0.5 lbs (200 g) per 50 ounces (1.5 liters) bottle which means that 800 bottles per year consumed by the average bottle-drinking household is equal to 350 lbs (160 kg) CO2 or the equivalent of driving 368 miles (592 km) with a car. Who would’ve thought something so minor and insignificant in our lives could create such a significant impact on climate change? Thus, we created ClimateSnap to bring you these stats with just one click!

What it does 🤔

Our website talks about the effects of climate change. It also has an object detection feature that scans the users’ webcam’s surroundings. While doing so, it returns the objects detected and outputs as a button that the user can click and get: information on how this object affects climate change, solutions to reducing its carbon footprint, and some statistics on its carbon emissions.

How we built it 🔧

Our entire project is a React App and is coded in mainly Javascript. Object detection implements TensorFlow object detection which is a computer vision technique that detects, locates, and traces an object from a still image or video. As well, it implements cocoSSD where the COCO dataset can be used to train object detection models. The dataset provides bounding box coordinates for 80 different types of objects, which can be used to train models to detect bounding boxes and classify objects in the images. Putting those together, we obtained a real-time output to detect objects in our surroundings. Furthermore, to provide the stats and info on the objects, we implemented the ChatGPT API. When the object is detected, the object is returned to a prompt that is sent to the ChatGPT API to get a response back.

Challenges we ran into 🏃

ChatGPT API implementation and time constraint!

Accomplishments that we're proud of 🥳

Implementing ChatGPT API for the first time! It was very useful and a big part of our project. We hope to implement it again in the future!

Designing a project with an amazing front end! A few of us who have done other hackathons in the past never had the opportunity to add an eye-catching front-end so to tie them together is a big achievement for all of us!

And of course, accomplishing all this in less than 24 hours!

What we learned 📚

Since we’re all relatively new to Javascript and React, we learned a lot more functions and implementation methods. As well, we’re always aiming to try out new APIs and so we learned how to implement two more today!

What's next for ClimateSnap ▶️

ClimateSnap hopes to generate more information with our prompts, allow for back-and-forth conversation with the selected objects for further learning, and provide other climate-related features such as a timeline visualizer of climate change.

Built With

Share this project: