Affekta

Real-time emotion recognition using facial expressions

Requirements

• Python
• OpenCV
• A facial expressions dataset (or use the default pretrained model)

Running Affekta

To run Affekta, browse to its main directory via the terminal and run affekta.py in sudo mode.

Unless altered, Affekta will use the pretrained model (default-model.xml) which was trained using the Cohn-Kanade AU-Coded Expression Database.

Training your own model

• Move your facial expressions image data to datasets/emotions/emotions_images
• Move the corresponding emotions data to datasets/emotions/emotions_data
• Run create_project.py to create the project structure
• Run extract_faces.py to preprocess and extract faces from your image data
• Run generate_model.py to train your model (saved to the models directory)

Using your own model

To use your own model, rename your model to defaul-model.xml, or alter the predefined model name in affekta.py.

Credits

This project is inspired by and based on the great work by Paul Van Gent.