Real-Time-Facial-Expression-Detection-Web-App
Real-time sentiment analysis of face using Keras and OpenCV with the help of web application based on Flask The web app is able to detect the following 7 emotions in real-time:
1.Happy
2.Angry
3.Disgust
4.Fear
5.Sad
6.Surprise
7.Neutral
The detection model used a combination of CNN layers followed by ANN layers, having TensorFlow.keras backend. Face detection is done via the CV2 library. The template for the web app is stored in templates folder as index.html We can also detect emotions of people in a recorded video by changing only a single line (line 11) in camera.py. (The line has been commented out in cmera.py itself for video usage)
Execution flow: Facial_Expression_Training.py (to build and store the model in .h5 format) ->model.py ->camera.py ->main.py