Skip to content

Repository files navigation

####################################################################################################
EMOTION LEVEL ANALYZER

Backend management consists of:

1. Openface toolkit manipulation - MMSA FET
2. EFA based AU seperation.
3. AU based K means clustering.
4. Emotion level Machine learning algorithms based on Age  - KNN, SVM, RF 
5. Models trained with Age variation. (pkl files included - RF is the best model)
6. Classifier predicting codes to test a video.

# Execute Program with the following command

python main.py --mode video --input \Video\S1_happy_1.mp4 --age 1

About

This is an Undergraduate research done with the objectives of Automatically Identifying the AU stimulation of children on their facial expressions, Development of AU based models to analyaze the varying emotion levels of children in videos and analysis and prediction of emotion levels based on varying age categories of children

Resources

Stars

0 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors