Vaak Eye alerts the store owners based on its scrutiny of over 100 characteristics of the subject that include its face, movement, clothes, etc
Japanese tech start-up Vaak has developed an artificial intelligence (AI) software that can track the possible shoplifters, by using footage from the security cameras. The software can detect restlessness, fidgeting and other possible suspicious body language.
Last year, the start-up helped to nab a shoplifter at a convenience store, where it had set up its software as a test case. The software picked up a previously undetected activity of shoplifting.
The software, called Vaak Eye, harnesses deep learning from over 100,000 hours of surveillance data. It alerts the store owners based on its scrutiny of over 100 characteristics of the subject that include its face, movement, clothes, direction, stride and even the purpose.
In addition, the software uses available macro data like crime incidence rate and weather condition to improve prediction. It also records each and every product the person has picked up and the time each item has been held by that person.
Before a possible shoplifting, Vaak Eye alerts the shop owners through a smartphone app.
In 2017, the global retail industry lost around US$ 34 billion in sales due to shoplifting – the largest source of shrinkage, as per a report by Tyco Retail Solutions.
There is a huge opportunity for such type of solutions. According to Gartner Inc, this year, retailers are estimated to invest US$ 200 billion in new technologies. They become more open to adopt technology in order to meet the need of their consumers, along with improving bottom lines.
Started in 2017, Vaak is presently testing the software in a few dozen stores in the Tokyo. This month, the start-up has started selling a market-ready version of the shoplifting-detection software. It is aiming to deploy the software in 100,000 stores all over Japan in three years.
The start-up has a funding of US$ 450,000 (50 million yen) from SoftBank Group’s AI fund. In addition, it is in the middle of series A round finding, seeking to raise 1 billion yen.