Social Distancing with Computer Vision

Over just a few months, more than seven billion lives were changed in ways that no one could have ever imagined. Economies came to a halt; global transportation and supply chains were forced to a standstill.

Like some parts of the world, businesses in Malaysia are gradually reopening under what our local government is calling the Conditional Movement Control Order. Essentially, we are allowed to venture beyond the confines of our homes and neighbourhoods, and are expected to obey guidelines such as wearing a mask, and practising social distancing, in order to control the risk of getting infected.

It’s common knowledge by now that social distancing is crucial in preventing the spread of COVID-19. By minimizing the amount of close contact we have with each other, we reduce our chances of contracting the novel coronavirus. Besides wearing face masks and eye protection, keeping at least one meter apart is the best way to cut the risk of COVID-19 infection, according to the largest review of studies to date on the transmission of the contagious disease. 

Stores are doing as much as they can to encourage social distancing. However, we’ve observed that such social distancing measures are only adhered to by the public as they are forced to queue up to purchase their goods.

After we move about freely, social distancing can easily be flouted unconsciously, especially within an enclosed space, when constant movement is deemed necessary to get your task done, for example, as you are doing your groceries. 

Imagine this, you are in a grocery store with 50 others, who are constantly in motion as they move from one aisle to another. While it’s also easy to get carried away with our thoughts and as we’re engrossed in the act of choosing and buying, we’re bound to flout social distancing rules whether we realize it or not.  

This lack of awareness could easily lead to a second wave of infections and relapses of public space closures.

How can AI help us systematically maintain social distancing in public spaces, especially enclosed ones such as grocers and retailers? 

AI enabled technology such as cashier-less checkouts and robots stock checkers were gaining traction from the  Retail Tech sector before the outbreak of the pandemic. However, as we embrace the new ‘normal’, many retailers are now operating at half capacity and the focus on driving efficiency becomes less about ensuring that shelves keep up with the demand, and more about ensuring that shoppers feel safe as they browse in stores. 

It is imperative for retailers to ensure that social distancing measures are in place from the time shoppers start lining up to enter their stores, to the moment they leave the store’s grounds. Although some retailers try to enforce such measures by putting guidelines on the floors of their stores with tape, these solutions can often be time-consuming and inefficient in the long run.  

Retailers looking at long-term and sustainable solutions to ensure the safety of their customers and employees need to begin looking at AI-enabled solutions. They would need to put in place processes, procedures, and solutions that can be automated with the help of technology. 

Computer Vision the solution to social distancing flouters

Since the start of COVID19, many AI companies have begun expanding their services to help curb the spread of the virus, with many leveraging its technology to meet new challenges faced by their clients. For example, many businesses are turning to technology for help in people tracking algorithms to monitor violations.

Computer Vision — a branch of AI technology also known as computer or machine vision –  comes in handy now more than ever, especially with the need for prevention in overcrowding and in determining if the occupancy threshold of a designated area has been exceeded. This allows retailers to avoid overcrowding by monitoring mounting traffic as it forms.

AI startups like Landing AI introduced an AI-enabled tool to help ensure individuals are keeping a safe distance from one another in the workplace. Established by one of the most prominent leaders in AI, Dr. Andrew Ng, Landing AI and its ‘social distancing detector’ analyses real-time video streams to help monitor and ensure that social distancing measures are met. The system will also be able to issue alerts to remind people to keep a safe distance if it finds anyone violating social distancing.

Another similar technology offered by Shapes AI, analyses camera feeds to detect if social distancing measures and the wearing of face masks are being observed. This allows business owners to receive real-time alerts and get access to data-driven actionable insights.

In addition to playing a part in ensuring the safety of shoppers, the insights obtained from AI technology can also help inform the public of how the measures are being adhered to, as well as when and where it is safe for them to go out.

Concerns about COVID-19 will linger for a significant amount of time. And until a vaccine becomes available, social distancing remains our best tool to help manage the pandemic, and reopen our economies. Relying on manual solutions only serve as a quick and temporary fix, and retailers will need to innovate in order to stay afloat in such volatile times.


Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

You May Also Like