Overview

Realtime Emotion Detection Using Keras https://youtu.be/DtBu1u5aBsc

Youtube Tutorial

https://youtu.be/DtBu1u5aBsc

Program is trained for 30 epochs and runs at 66% accuracy.

Dependencies:

• pip install numpy
• pip install pandas
• pip install tensorflow
• pip install keras
• pip install opencv-python

Download haarcascades file from here=> https://github.com/opencv/opencv/blob/master/data/haarcascades/haarcascade_frontalface_default.xml

Note: Kaggle dataset can be downloaded from:

http://www.mediafire.com/folder/trbjv7bysiycl/challenges-in-representation-learning-facial-expression-recognition-challenge

Usage:

First Run Emotion_recognition.py => python Emotion_recognition.py
Then run vidotester.py

Acknowledgements:

• http://sefiks.com/2018/01/01/facial-expression-recognition-with-keras/
• https://medium.com/@jsflo.dev/training-a-tensorflow-model-to-recognize-emotions-a20c3bcd6468