Face and Emotion Recognition
The following demo introduces face detection and emotion recognition with deep learning.
1. Face detection (OpenCV Haar Cascade) is to find the face or faces in an image.
2. Emotion recognition is to identify the emotion shown by a face.
Emotion classification model: CNN mini Xception
Classes: angry / disgusted / scared / happy / sad / surprised / neutral
After click on button “OK!“, your browser will ask for permission to use your camera.
If your camera setup is correct and permission is granted successfully, the camera’s live video feed can be observed in the 1st picture.
2. Press the button “Take Photo” which will take a photo of yourself with your camera, which can be seen on 2nd picture.
Please make sure your face is in the picture!
3. Push the button “Process image“, then AI will draw a rectangle around your face and tries to figure out your emotion (happy, sad, scared, angry, neutral) and the probability of the guess on 3rd picture.