Your face tells a story

gzlaser

NOT even Dan Brown and his Da Vinci code-breakers dared deal with the mystery of Mona Lisa's smile. But Nicu Sebe, a computer expert at the University of Amsterdam, the Netherlands, did.

He used \"emotion recognition\" software to process the famous enigmatic portrait and found Mona Lisa happy (83 per cent) and slightly disgusted (9 per cent).

Faces reveal emotions, psychology, computer science, and engineering researchers are joining forces to teach machines to read expressions. If they succeed, your computer may one day \"read\" your mood and co-operate. Machines equipped with emotional skills could also be used in teaching, robotics, gaming, security, and for psychological diagnoses.

Sebe doesn't spend research time analyzing famous images - that's just for fun. Anyway, calling Mona Lisa \"happy\" is not accurate science. Why? Because to read an emotional state, a computer needs to analyze expression changes against a neutral face, which Da Vinci did not provide.

If that's the case, are computers even close to reading emotions? You bet.

\"Mind Reader\", a system developed by Rosalind Picard at the Massachusetts Institute of Technology in the US, uses input from a video camera to do real-time analysis of facial expressions. It reports on whether you seem \"interested\" or \"agreeable\" or if you're \"confused\".

The system can help people decode others' emotions. Picard says this means we could teach a machine to be as sensitive as a human.

In fact, a machine can be even smarter than people since it can tell if a person is lying or just \"performing\" by analyzing one's facial movements.

Jeffrey Cohn, a psychologist at the University of Pittsburgh, uses the Facial Action Coding System to decode human emotions. The system classifies more than 40 action units (AUs) of the face to tell people's real emotions.

\"The face is always visible,\" Cohn says. \"eople communicate a lot about feelings and thoughts through facial expression.\"

Cohn is working to get machines to read AUs and describe which muscles moved and how. Such applications could be used in lie detection.

Cohn studied a videotape of a criminal who professed to be sad about the murder of several family members and tried to pin the blame on someone else. But Cohn saw no genuine sadness in the woman's face.

Sadness is a combination of AUs that is difficult to do voluntarily. It involves pulling down the corners of your lips while bringing your eyebrows together and raising them. What the subject did was raise her cheeks to simulate the lip curl. Her brows stayed smooth.

This means, even though your mouth lies, your face doesn't, and the machine will know it all.