Search | Contact | Subscribe | Sign In or Register

Crisis Response Journal Crisis Response Journal

Software can read subliminal facial expressions to decipher ‘deeper’ human personality traits in real time

Posted on 19th September 2017 at 20:36pm

Sophisticated technology can tell – live – if people are lying, disagreeing, curious, judgmental, passionate or nervous with high levels of accuracy, says developer, Human.

The company, a pioneer in Artificial Emotional Intelligence (AEI), has developed new software that can read subliminal facial expressions and convert these into a range of emotions and specific characteristic traits in real time. The software takes AEI and facial recognition a step further into the dynamic realms of characteristic detection.  

Going deeper than simply deciphering human emotion, Human’s patent pending software goes beyond using partial facial recognition, robust camera angles and state-of-the-art pixelated raw data to reveal typical personality traits. 

Human says the technology provides advanced levels of unbiased insight to help companies better predict opportunities, manage risks and make wiser decisions in a variety of situations, ranging from employing the right person and cultural fit for the job, to supporting fraud detection, providing professional sports scouting intelligence and more.

The innovative software minimises human bias, is cost effective and time efficient. Yi Xu, CEO and Founder at Human explains: “Understanding what makes someone passionate, nervous, dishonest or curious for example has, to date, been about the human subconscious biased view, which has always been a gamble.

“Our software adds another layer of certainty to simple gut instinct with supporting data. Using the power of deeper-rooted technology that not only looks at how someone is feeling at that point, but also at what kind of person they are, will allow us much greater insight into individual human behavioural traits, which can be used to great advantage when making better, more informed decisions.”

Human says it can take security and intelligence insight to another level, something that previously, has been almost impossible to achieve. Through the deciphering of live characteristic traits and emotions utilising deep learning, it says the security and intelligence world will be better equipped data that will keep both the security provider and its customers safe. “This will allow them to make vital decisions, such as monitoring extreme, or even life-threatening behaviour in some cases,” Yi Xu told CRJ.

With the capability to provide this technology across multiple platforms, including mobile, web and API delivery, Human can analyse past and live video footage with no additional equipment needed. This technology helps to support suicide prevention, face detection and verification among crowds, extreme behaviour monitoring, societal disturbances and monitoring terrorist activity via prevention.

“The ongoing scanning of emotions and characteristics in both small and large crowds has provided our clients with the ability to flag any extreme highs and lows in emotions,” continued Yi Xu. “When narrowed down to extreme happiness, sadness, nervousness among others, it is giving both general and security staff the foresight and ability to step in before the worst case scenario presents itself. This continues to aid the ongoing pattern-spotting and monitoring that exists today.”

The flagging of characteristics and emotions, combined with face detection, allows for crowd monitoring and tracking to be taken from a previously reactive to a preventative state. “Businesses now know they are acting with the added benefit of psychological and emotional data, as well as a visual track,” the CEO added. “This benefits detecting of wanted or suspected individuals as well as identifying people who may be about to commit an offence.”

Using raw video footage, Human’s technology is able to tell live in real time what a person is feeling, as well as provide insight into certain personality traits, which can be used to predict human behaviour, as this video shows. Check out the results of the video here  

Share Your Thoughts
Sign In or Register to leave a comment
Back to R & D Back to Top