It helps me find vintage treasures, encourages spending less money on second-hand clothes, and helps the planet!
However, the market lacks an efficient thrifting platform for mainstream visibility. The time-consuming process of searching through hundreds of clothes poses as a barrier to thrifting. Therefore, this is a solution to marry the humble beauty of thrift clothes and the power of technology.
Meet Quark, a personalized thrifting platform aimed to bring together online consumers with physical thrifting stores!
- Armed with a Convolutional Neural Network and intelligent classification technology, this application allows thrift stores to simply upload pictures of their products in seconds, leaving the program to handle labelling and marketing.
- At the same time, Quark users are met with the thousands of small online stores and physical in-person stores offering a variety of products, which are all filtered through the user's interests and previous purchasing habits.
Note: This project currently only serves as a clothing classification and color generator platform. More features to come!
In other words, it's a full-stack application designed in responsive (mobile-friendly) practices and lightweight server processing
To get a local copy up and running, follow these simple steps!
- Clone the repo
git clone https://github.com/yccgabby/quark.git
- install all necessary packages
pip install -r requirements.txt
cd frontend
npm install
- run the backend (make sure you're in root project directory)
flask run
- run the frontend
cd frontend
npm start
- Upload a picture of a piece of clothing to the upload tab of the app, and the program will
- try to classify the type of clothing it is based on a pre-trained Tensorflow CNN
- generate a five-color palette that complements the article of clothing
Gabby Chan - a current systems design engineering student at the University of Waterloo
Feel free to email me at g44chan @ uwaterloo dot ca
Project link: http://github.com/yccgabby/quark.git