Hundreds of retail stores in UK now use visual AI

This year, a few of UK’s largest retailers have successfully deployed AI-powered CCTV surveillance across hundreds of their stores. The foundation for this technology upgrade was provided by SAI Group’s visual AI platform, which can automatically detect a wide variety of theft scenarios such as aisle theft, item concealment, item switching, and checkout fraud.

Co-op faces over 1000 crimes every day

The notion that shoplifting and retail crime is victimless is not true, according to Paul Gerrard of Co-op. Every day in Co-op, over a thousand crimes were reported, with 120 store staff facing risks of serious abuse or threat and 4 of them facing attacks, often with weapons.

Retail stores in a race against time to deploy AI

In what would be the first public demonstration of a new AI system being installed by retailers in time for Christmas, Co-op is in a race against time to deploy the solution across their entire chain of stores. The AI solution detects several theft scenarios such as when a suspected shoplifter puts items into a bag.

Huge rise in shoplifting and fall in police attendance rates

Shoplifting has skyrocketed over the last few years and the retail industry has lost over a billion pounds until now. At the same time, police attendance rates have more than halved since 2016 to date, and the industry has been lobbying for improved and responsive policing. The police force cites several reasons such as lower value of theft, online or remote means of carrying out investigations, and lack of credible evidence for the reduced attendance rates (see news).

Video evidence in focus

The UK government’s Project Pegasus, which is envisioned as a digital platform, will enable the retail industry to share evidence of retail theft, enabling police to initiate swift action to identify the offenders and bring them to justice. The government also intends to establish a specialized police team with the objective of building intelligence on organized crime in retail across the country.

At the heart of Project Pegasus is the requirement of retail stores to submit CCTV recording of entire shoplifting incidents along with a still image of the offender, which can be used by the police to run the image through facial recognition technology. The swift identification of the offenders and the quick response ability to act by law enforcement can restore the confidence among the retail store staff.

Challenges in CCTV footage submission

Incident reporting with legally admissible evidence can prove to be a huge burden on the store staff. They need to wade through hours of CCTV footage to identify and extract video clips pertaining to the crime. And when multiple such events take place across multiple stores of a retail chain, the productivity loss for the retail chain can add to the financial losses they already face from retail shrinkage.

Automated evidence collection

SAI Group’s visual AI solution works with existing CCTV installations, enabling real-time notifications about suspicious events within seconds. The solution works in the background, analyzing and shortlisting footage of interest, discreetly informing the store staff about incidents as and when they occur, and providing ready to submit, legally admissible video and photographic evidence – without any manual intervention.

Easy to deploy, cost-effective AI

One of the biggest challenges that AI solution providers face today is to make their solutions cost-effective. Whether it is a fixed cost or a per-seat subscription, AI solutions need to ensure that:

• They cover their costs in terms of the computing resources that they consume
• Their customers do not find it difficult to justify the costs incurred while implementing AI solutions.

SAI Group offers highly competitive solutions that ensure return on investment within a few months of deployment.

About SAI

As a leader in computer vision technology, SAI Group delivers cutting-edge, multi-modal AI solutions into retail environments. Using a unique platform approach, its technology uses existing camera systems to target losses, increase store safety, and underpin operational efficiencies.

All solutions are built from the ground up to ensure the highest levels of security and data protection, respecting the privacy expectations of the public and operating to stringent ethical standards while delivering substantial value to our clients. Globally, SAI monitors millions of transactions per day, protecting the revenues from tens of millions of product sales and hundreds of millions of customer interactions. Its models also accurately identify anti-social behaviour, aggression and violence, helping to de-escalate situations with real-time interfaces to security officers and operations centres.