chonkii

Inspiration

Our main motivation for creating this program stems from the fact that cat obesity remains a prevalent issue in America, and has continued to worsen in recent years.

We first identify five health risks associated with feline obesity:

Obese cats overall have a worse quality of life and a significantly lower life expectancy.
Obesity leads to immune system complications, making them more prone to infection.
Liver failure.
Difficulty for cats to groom themselves leads to skin problems.
Increased risk of diabetes.

What it does

The user uploads an image of their cat, which is then ranked on our 3-way classification scale (not-chonky, kinda-chonky, or chonky). The model outputs:

The classification (whether the input is a cat or not)
The chonky rating (is the cat chonky, kinda-chonky, or not-chonky)
A numerical score of the chonky rating

How we built it

We used React to handle the file uploading, imported a TensorFlow model for object detection to classify how 'chonky' the cat is, and rendered said data.

Challenges we ran into

Integrating the cat detection model built-in Python with JavaScript
Learning a completely new front-end JavaScript library, React.js

Accomplishments that we're proud of

We are proud of the UI we designed and with being able to deploy an application in time.

What we learned

Front end development in React.js
Styling with CSS
Deploying a website
Data formatting with JSON
Version control with GitHub

What's next for chonkii

In the future, we want to implement a way to provide recommendations on optimal diets and exercise routines for chonkier cats can be customized depending on the category the cat is placed into. Additionally, we would like to develop a mobile application, similar to Apple's Health App, that syncs and monitors the cat's chonk, with daily notifications to remind the user when and how much to feed their cat. This way, the user is able to monitor their fluffy friend's health, all at the push of a button.

Created by Danny Do, Anish Ganti, Nadia Rodriguez, Sajiv Saksena, and Relena Lai