New Computer Vision System: Emotion Recognition

fake or real

Everyone, in some point or the other in life, must have faked of being in pain or being ill, to escape from going school, college, office or so on. We humans are not always correct to distinguish a fake emotion from the genuine one. Lately, researchers have developed a computer system that can easily distinguish when an individual is genuinely in pain or is just faking it away more accurately than the humans. 

A team of researchers from the University of California, San Diego and the University of Toronto develops the computer system based on the Facial Action Coding System. The system can identify the various pattern and vital features of expression of the person’s face that go unnoticed by humans. Humans face can express a variety of information such as pain, smile, sadness, happiness and many others.

Brain can stimulate fake expression

Researchers observed that the humans performed fairly badly when asked to point the real expression from the faked one, even when trained for the same, the accuracy increased to 55 percent. On the other hand, computer system attained 85 percent of accuracy in the same test by identifying the slight difference between the voluntary and involuntary expression and facial movement.

Kang Lee from the University of Toronto said that the human brain can easily simulate and fake such expression, even when it is not actually experiencing it, thus easily convincing other people.

Researchers say that when a person is in actual pain, the facial muscles contract in a definite pattern and people try to use same facial muscles in an attempt to imitate the real pain. The research team says that emotions, whether voluntary or involuntary, have different paths in the brain. The brainstem and the spinal cord are involved in expressing involuntary emotions, whereas voluntary expression depends on motor planning area of the cerebral cortex and require more conscious attention on behalf of the person.

Genuine Expression versus Fake Emotion

To differentiate between a genuine from a deliberate expression, one needs to observe the dynamics like how often, rapidly muscles move, how much do they move and when do they move. The minute changes in the facial muscles get unnoticed by humans and therefore makes it difficult to distinguish.

According to a study, the feature that conveys the truth in the mouth and by closely observing how and when it opens can help a lot. A person, who is faking a pain, opens mouth too often and without much variation. The researchers observed that the duration for which mouth remain open and the period between the opening of the mouth is too regular, unlike as observed during a real pain. Further research confirmed such consistency is an expression associated with fake expressions.

As per the author of the research, Marian Bartlett, who is also a professor at UC San Diego’s Institute for Neural Computation, the computer vision system can be used to identify cases where human facial expression may furnish vital clues related to emotion, health, thought, physiology and even in situations such as detecting whether a driver is felling dizziness or students are paying due attention and to determine the effectiveness of the treatment on certain disorders, so on.

In the future, such system will also help in revealing the truth or level of pain if someone is trying to mask it, while interrogating a suspect. But be aware that the same is being not used by your boss, when you try to fake a pain for a leave and plans for a holiday.

Explore further

Leave a Comment