Opensource deep learning framework TensorFlow is used in Facial Expression Recognition(FER). The trained models achieved 65% accuracy in fer2013. If you like this, please give me a star.
FER requires:
Only tested in Ubuntu and macOS Sierra. Other platforms are not sure work well. When problems meet, open an issue, I'll do my best to solve that.
You will have to download the pre-trained models here. Then run the demo that Detecting the face(s) in video captured by webcamera, and recognize the expression(s) in real-time.
python3 main.py demo
And you can add --showbox
option to show box of faces.
You can train models by yourself. Download the fer2013 datasets in kaggle(91.97MB).
Extract the data to data/fer2013
folder.
Then train model.
python3 main.py train
If any issues and suggestions to me, you can create an issue.