Emotion classification

real-time face detection and emotion classification

• the test accuracy is 66% in the kaggle competion dataset fer2013
• the test accuracy is 99.87% in the CK+ dataset
• the average emotion classifer model predict cost time is 4~ 10ms in real-time videos captured from webcam ##dataset source:
• CK+:

The Extended Cohn-Kanade (CK+) database distribution.there are 593 sequences across 123 subjects which are FACS coded at the peak frame ONLY 327 of the 593 sequences have emotion sequences.(emoton labels.{0: 'neutral', 1: 'anger', 2: 'contempt', 3: 'disgust', 4: 'fear', 5:'happy', 6:'sadness', 7:'surprise'})

• fer2013:

the kaggle competitions: Challenges in Representation Learning: Facial Expression Recognition Challenge https://www.kaggle.com/c/challenges-in-representation-learning-facial-expression-recognition-challenge The dataset provided in the competion consists of gray scale images which are 48 x 48 in dimension and the corresponding labels consisting of 7 emotions. (emtion labels.{0:'angry',1:'disgust',2:'fear',3:'happy', 4:'sad',5:'surprise',6:'neutral'})

classification model use mini_XCEPTION

Emtion classification examples

• test image

• test video

Reference

• Xception: Deep Learning with Depthwise Separable Convolutions
• Keras Applications

Prerequisites

• Python 3.5
• OpenCV
• Dlib
• keras
• Numpy
• TensorFlow