Facial expression classification using salient pattern driven integrated geometric and textual features
- Using CK+ face dataset
- Extract salient patterns of geometric and textual features
- Based on the above features, using one vs rest SVM and other traditional machine learning algorithms in traditional_modeling.py
- Also apply on deep learning frameworks like CNN
- The accuracy of SVM has reached 86.58% for 7 emotions classification and more than 96% for 6 emotions classification (except contempt)