Facial Emotion Recognition Project 🚀

ML based project written in python convolutional neural network architecture and trained on the FER2013 dataset with FERPlus labels.

✨By Ayush Kumar Sahu, Akash Kumar Verma, Deepanshu Pawar & Amit Dewangan✨
Star🌟 this repository and fork🍴 to save it to your repos for easy access.

Connect with us :

LinkedIn-Ayush-Sahu LinkedIn-Akash-Verma

Twitter

What is FER or facial emotion recognition?

It is the process of using computer algorithms to identify and analyze emotional states of a person by analyzing their facial expressions.

FER technology is used in various fields such as psychology, neuroscience, marketing, and human-computer interaction to better understand the emotions and behaviors of individuals.

The technology works by analyzing facial features such as eyebrows, eyes, nose, mouth, and cheeks to detect patterns of muscle movement that correspond to different emotions.

Built With

Keras
Tensorflow
OpenCV

Prerequisites (install all technologies used)

dependencies can be installed using pip package manager for python

python >= 3.7.9
keras >= 2.4.3
tensorflow >= 2.3.1
opencv >= 4.4
sklearn >= 0.23
numpy >= 1.18.5
pandas >= 1.1.2
matplotlib >= 3.3.1

Installation - Running the project in your environment

Open terminal on the project folder

Clone the repo

git clone https://github.com/Akash-095/Minor-FER

Change to Minor-FER directory

cd Minor-FER

Install required packages

Install all dependencies as stated above, read individual technology documentation for further guide to install

Usage

To train the model use the following command

 python fer.py

The model can make predictions on saved images by providing the image path using the following command

 python img_predict.py img_name.png

It can also predict on saved videos

  python vid_predict.py vid_name.mp4

Or by using a live camera

  python live_cam_predict.py

Built with ❤️ by the maintainers