Automatic Micro Expression Recognition and Analysis

 Automatic Micro Expression Recognition and Analysis

Introduction:

With the advancement in technology and the machine learning methods various automatic processes are possible to predict human behavior and his emotions. Micro Expression Recognition as addressed in this research is also a machine dependent process which can reveal the hidden information and the intentions behind a micro-expression. These micro expressions of human beings which are otherwise very deceitful and hard to predict can be automatically monitored and the meaning behind them can be predicted with machine learning techniques. This technological advancement allows to uncover various hidden intentions and the affiliated emotions with micro expressions as to predict better about human beings. The research will focus on the machine learning methods and use of Spontaneous Micro Expression Database (SMIC) for the prediction of hidden emotions and revealing the truth behind the micro expressions.

Problem Statement:

Most of the micro expressions are hard to see with naked eye therefore there is an immense need to indulge technological methods and SMIC with advanced Machine Learning Techniques for uncovering the hidden emotions and revealing the truth. Advanced micro expression recognition still faces some lags, this research will be a huge contribution to fill these gaps and proceed in the domain of micro expression recognition.

Research Question:

The involuntary facial expressions for revealing the hidden emotions will be the core question to this research. Evaluation and comparison of the existing techniques for micro expressions recognition and contribution in the existing research to contribute to the field of Artificial Intelligence and Advanced Machine Learning Methods.

Literature Review:

The term micro expression was first introduced by Issacs and Haggard in 1966 [1]. Three years later, it was reported by Friesen and Ekman that micro expressions existed in the independent work that they were performing [2]. They were working on a film of psychiatric patient who revealed later that her plan was to commit suicide. The hidden anguished look in the happy emotions of the lady were revealed when the video was watched later under careful consideration and controlled environment. Ekman also indicated that lies can be detected with the help of micro expressions if they are observed carefully [3][4]. Cheating and lying can be easily detected with correct use of technology in the field of micro expression and their recognition for the hidden emotions. The Micro Expression Training Tool (METT) that was developed by Ekman was a major step towards improvement of micro expression detection [5][6]. Ekman and Friesen developed the FACS (Facial Action Coding System) for the description of Action Units (AU) in facial expressions [7]. The research performed for the recognition of various emotions like surprise, anger, sadness, and fear were the source of prototypical emotional expressions [8][9][10]. Gentleboost and SVM classifier was used by Wu et al. for the recognition of Micro Expressions using METT Tool [11].

Contribution:

The research of ME recognition was done by various researchers at different times in history, but the current research will focus on the automation of this research by proposing efficient use of technology and advanced mathematical techniques for the detection and recognition of micro expressions. The positive advancement is possible in the of ME recognition by the successful implementation of this research. 

Bibliography:

[1] E. Haggard and K. Isaacs, “Micromomentary facial expressions as indicators of ego mechanisms in psychotherapy,” Methods of research in psychotherapy. New York: Appleton-Century-Crofts, pp. 154–165, 1966.

[2] P. Ekman and W. Friesen, “Nonverbal leakage and clues to deception,” DTIC Document, Tech. Rep., 1969.

[3] P. Ekman and M. O’Sullivan, “Who can catch a liar?” American psychologist, vol. 46, no. 9, p. 913, 1991.

[4] P. Ekman, “Darwin, deception, and facial expression,” Annals of the New York Academy of Sciences, vol. 1000, no. 1, pp. 205–221, 2003.

[5] “Microexpression training tool (METT),” San Francisco: University of California, 2002.

[6] “METT micro expression training tool,” CD-ROM. Oakland, 2003

[7] P. Ekman and W.V. Friesen, The Facial Action Coding System: A Technique for the Measurement of Facial Movement. Consulting Psychologists Press, 1978.

[8] M. Yeasin, B. Bullot, and R. Sharma, “From Facial Expression to Level of Interest: A Spatio-Temporal Approach,” Proc. Conf. Computer Vision and Pattern Recognition, pp. 922-927, 2004.

[9] S.P. Aleksic and K.A. Katsaggelos, “Automatic Facial Expression Recognition Using Facial Animation Parameters and Multi-Stream HMMS,” IEEE Trans. Signal Processing, Supplement on Secure Media, 2005.

[10] I. Cohen, N. Sebe, A. Garg, L.S. Chen, and T.S. Huang, “Facial Expression Recognition from Video Sequences: Temporal and Static Modeling,” Computer Vision and Image Understanding, vol. 91, pp. 160-187, 2003.

[11] Q. Wu, X. Shen, and X. Fu, “The machine knows what you are hiding: An automatic micro-expression recognition system,” in Proc. 4th Int. Conf. Affective Comput. Intell. Interaction, 2011, pp. 152–162.