US tests $3.5m computerised lie detector

The US Department of Homeland Security has earmarked $3.5m to help scientists develop next-generation computer-based lie detectors.

The funds will be awarded to computer scientists at Rutgers, The State University of New Jersey, who are already researching how subtle body movements, such as shoulder shrugging, hand gestures or slight changes in facial expression, may indicate deception.

The goal is to capture these non-verbal cues on camera, have a computer analyse them and deliver immediate input on the subject's likely truthfulness.

Researchers foresee their techniques helping immigration officers screen people more quickly and accurately at border crossings and ports of entry into the US. The techniques could also bolster security for buildings such as embassies, and help law enforcement personnel in routine interrogations.

"Looking for what we call 'micro-expressions' and 'micro-gestures' associated with deception would be a major leap over today's polygraph technology, which is time consuming and requires professional administration," said Dimitris Metaxas, computer science professor and director of the university's Center for Computational Biomedicine Imaging and Modeling.

"Even under the most controlled conditions, lie detector tests based on body physiology are roughly 50 per cent reliable. But we believe gestures and expressions are a lot harder for someone to mask, and do not vary significantly among races and cultures.

"Micro-expressions may easily escape notice by human observers, but can be reliably picked up on camera and quickly detected by computer, giving interrogators new tools to do their jobs confidently."

Metaxas explained that he is building on his earlier research into the computer modelling of facial expressions and the dynamics of body organs, such as a beating heart or blood flowing through vessels.

This work employs image-based models that enable computers realistically to simulate subtle yet detailed muscular movements, such as those that make up facial expressions.

However, the scientists pointed out that "several challenges remain" including accurately correlating facial expressions to deceptive intent and verifying how consistently these occur across cultures.

Another problem centres on the incorporation of new cues, such as gestures and body postures, into the computer models.

The researchers will also study how accurately they can capture movements under 'real world' conditions of varied lighting, background motion and potential distractions such as people wearing glasses and men with facial hair.

While Metaxas will conduct early studies with multiple cameras, his goal is to make the recognition and analysis procedures robust enough to work with a single camera.

In addition to using cameras to capture images, the team will employ 3D sensor technology to capture the range of body movements. Rutgers is collaborating with Lockheed Martin on 3D sensor development and integration.