Facial Expression Recognition using Keras in Python

By Kushagra Kinjawdekar

Using webcam as a live source, detecting your face expression, having trained on the kaggle facial expression dataset using Tensorflow, Keras, convolution neural networks in Python.

REQUIREMENTS:

If you want to train the model by yourself, then get the dataset through kaggle from below link:

Facial Expression Dataset

1. Tensorflow - 2.3

2. Python - 3.6

3. Keras - 2.2.4

Additional Python libraries need to be installed (through pip if you don't have.. ) are:

1. Utils ( pip install utils)

2. Livelossplot (pip install livelossplot)

 

USAGE :

1. If you are using anaconda prompt first load your environment and go to your library where the project is stored by using the command 

   cd "C:\Users\......" .

2. Then open main.py using python command :

     python main.py and hit enter. The code will compile and get executed.

3. Now go to your default web browser and open http://127.0.0.1:5000/ (for window user ) , for Linux you can check your local host Id.

(Optional) : For those who are training their own model

1. Train your data and download the weights obtained from it.

2. Only change the weights to your own weights in the camera.py file currently (model_weights.h5) and Done.

EXAMPLE: