Neural network technology for pattern recognition, stock prediction and market forecasting

EigenExpressions for Facial Expression Recognition

Download now Matlab source code
Requirements: Matlab, Matlab Image Processing Toolbox, Matlab Neural Network Toolbox.

We propose an algorithm for facial expression recognition which can classify the given image into one of the seven basic facial expression categories (happiness, sadness, fear, surprise, anger, disgust and neutral). PCA is used for dimensionality reduction in input data while retaining those characteristics of the data set that contribute most to its variance, by keeping lower-order principal components and ignoring higher-order ones. Such low-order components contain the "most important" aspects of the data. The extracted feature vectors in the reduced space are used to train the supervised Neural Network classifier.

This approach results extremely powerful because it does not require the detection of any reference point or node grid. The proposed method is fast and can be used for real-time applications. This code has been tested using the JAFFE Database, available at http://www.kasrl.org/jaffe.html. Using 150 images randomly selected for training and 63 images for testing, without any overlapping, we obtain an excellent recognition rate greater than 83%. The semantic data ratings for this database are available at http://www.kasrl.org/jaffe_info.txt.

Index Terms: Matlab, source, code, facial, expression, recognition, JAFFE, neural networks, PCA, network, expressions, face, principal component analysis.

Release 1.0 Date 2007.02.22
Major features:

• Facial expression recognition based on Neural Networks
• PCA-based dimensionality reduction
• High recognition rate
• Intuitive GUI
• Easy C/C++ implementations
• Demo code (protected P-files) available for performance evaluation