Repository files navigation

This project aims to classify the emotion on a person's face into one ofseven categories, using deep convolutional neural networks. This repository is an implementation ofthis research paper. The model is trained on theFER-2013 dataset which was published on International Conference on Machine Learning (ICML). This dataset consists of 35887 grayscale, 48x48 sized face images withseven emotions - angry, disgusted, fearful, happy, neutral, sad and surprised.

• Python 3,OpenCV,Tensorflow
• To install the required packages, runpip install -r requirements.txt.

The repository is currently compatible withtensorflow-2.0 and makes use of the Keras API using thetensorflow.keras library.

• Clone the repository

git clone https://github.com/saranshbht/Emotion-detection.git

• This implementation by default detects emotions on all faces in the webcam feed. With a simple 4-layer CNN, the test accuracy reached 63.2% in 50 epochs.

• First, thehaar cascade method is used to detect faces in each frame of the webcam feed.

• The region of image containing the face is resized to48x48 and is passed as input to the CNN.

• The network outputs a list ofsoftmax scores for the seven classes of emotions.

• The emotion with maximum score is displayed on the screen.

• "Challenges in Representation Learning: A report on three machine learning contests." I Goodfellow, D Erhan, PL Carrier, A Courville, M Mirza, BHamner, W Cukierski, Y Tang, DH Lee, Y Zhou, C Ramaiah, F Feng, R Li,
  X Wang, D Athanasakis, J Shawe-Taylor, M Milakov, J Park, R Ionescu,M Popescu, C Grozea, J Bergstra, J Xie, L Romaszko, B Xu, Z Chuang, andY. Bengio. arXiv 2013.