Automatic Facial Experssion Analysis

Supervisor: Ke CHEN
Difficulty grading: SH/INF=U, BM=N, CM=C
Area:    Artificial Intelligence
Max number of students who can do this project: 2

As a salient feature of human beings, emotions play a critical role in our daily life, e.g. perception, human interaction,rational decision-making, and human intelligence. As one of most important aspects in emotions, facial experssion conveys rich emotional information during interpersonal communications. The project is planed to develop an automatic facial emotion analysis system based on recent pattern recognition techniques.
In general, automatic facial experssion analysis consists of two non-trivial tasks: extraction of salient features from a facial image to form an individual-independent representation for characterising emtional states and detection/recognition of emotional states. While a complete prototype can be developed, the topic might be decomposed into two separate projects; one focuses on the feature extraction and the other mainly looks into the effective techniques for detection/recognition of emotional states from a facial image or seqeunce. In terms of the definition of a project, a prototype of automatic facial experssion analysis is developed as deliverable and the GUI is essential to such a prototype as well.


  1. B. Fasel and J. Luettin, "Automatic facial expression analysis: A survey," Pattern Recognition, vol. 36, no. 2, 2003, pp. 259-275.
  2. A. K. Jain, R. P. W. Duin and J. Mao, "Statistical pattern recognition: A review," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 22, no. 1, 2000, pp. 4-37. (comments: a good survey for underpining techniques)

COURSE PREREQUISITES: Machine Learning (Pattern Recognition), Image Processing

EQUIPMENT: PC, appropriate I/O device, Matlab (ideal with the image processing toolkit)