lltlien / virtual-hair-dye-try-on-web-application

Source code for a web application developed as the final project for the Computational Thinking course (CS117) at the University of Information Technology (UIT)

Virtual Hair Dye Try-on Web Application

In recent years, there has been a growing trend in the use of hair dye as a means of exploring various styles and transforming one's appearance. With the continuous advancement of technology, a new trend has emerged - virtual hair dye. The primary focus of this project is to introduce a solution for virtual hair dye, with emphasis on the key processes of hair segmentation and recoloring. The proposed solution leverages advanced algorithms and image processing techniques to identify and isolate the hair region in an image. This allows users to preview and apply different hair colors with ease, without the risk or commitment associated with traditional hair dyeing methods. The virtual hair dye technology enables individuals to discover their ideal hair color without any adverse effects.

Web App Demo

Installation

git clone https://github.com/lltlien/Virtual-Hair-Dye-Try-on-Web-Application.git
cd Virtual-Hair-Dye-Try-on-Web-Application

Requirements

Flask==2.2.2
numpy==1.24.3
opencv_contrib_python==4.5.5.62
Pillow==9.4.0
scikit_image==0.19.3
torch==2.0.1
torchvision==0.15.2

Ensure you have Node.js installed on your system.

Install Node.js

Quick Usage

First cd into the frontend folder, install all dependicies and run frontend.

cd frontend
npm install
npm start

Create virtual environment:

py -m venv env
env\Scripts\activate
pip install -r requirements.txt

To run backend:

py app.py

Once both frontend and backend are running, open http://localhost:3000 in your browser to access the application.