This website uses cookies. By using this site, you are consenting to the use of cookies on your device. LEARN MORE OK

The Computer Vision Revolution Is Here. Are You Ready?


Computer vision has opened up many possibilities across industries.
Companies can leverage computer vision to drive more effective marketing efforts and higher profits

In recent years, tech companies have made major advances in artificial intelligence and computer vision to create new business models and transform industries. Examples such as Google’s Waymo self-driving cars, Tesla’s autopilot mode, the Amazon Go store, Apple’s Animoji, and Amazon’s Virtual Mirror and EchoLook save consumers time and improve their overall experience, but this technology doesn’t have to be limited to tech giants. Industries like airlines and hotels, too, can harness the power of computer vision to improve their own customer experiences.

Computer vision technology has advanced significantly over the last few years. Early successes in AI surfaced in 2012, when deep learning algorithms recognized complex shapes in a computer vision competition. By 2015, another computer vision algorithm scored a 0.2% lower error rate than humans at identifying hundreds of object categories and millions of images. In 2016, Google AI made significant advances in the detection of diabetic retinopathy, a leading cause of blindness among adults, and the technology is expected to keep improving due to the sheer number of images that its algorithms learn from. In 2018, AI-enabled diabetic retinopathy detection devices were approved by the FDA, confirming that computer vision algorithms can do as well as or, in certain cases, even better than humans.

As the research community shares its knowledge, smaller companies can use cameras, sensors and the internet of things to similarly make meaningful customer experience improvements. For example, computer vision can help you measure customer happiness by using cameras to show how happy customers are when they come out of your store, or help you measure customer preferences, and target marketing offers to them based on what they’re buying or how they’re feeling.

So how should you start? As Andrew Ng, computer scientist and one of the world’s best known AI experts, says, “Pretty much anything that a normal person can do in less than one second, we can now automate with AI.” Here are two ways that companies can leverage computer vision:

  1. Cognitive complexity: There are two levels of cognitive complexity: Recognition, the one-second tasks that humans can perform, and reasoning, a series of much smaller tasks that help us make decisions. Casino player skill assessment is one example of reasoning, where many small factors such as player bets and player moves need to be understood and reasoned together to determine a player’s skill level.

    For example, when a player sits down at a slot machine, the casino has complete information about that player’s average wager, number of wagers and length of play. However, if a player chooses to play at the table, the casino must rely on its dealers’ and supervisors’ observations to estimate the player’s value to the casino, which can be wildly inaccurate. Every year, casinos spend hundreds of millions of dollars to measure this, since overestimating a player’s value destroys the lifetime profitability for the casino, while underestimating it negatively affects the player’s loyalty to the casino because they could miss sending the players relevant coupons or offers to entice them to play again.

    Computer vision can take human error out of the equation by shooting video of game play on a sample blackjack table. By using an automated cloud system to analyze player behavior in the videos, such as how much they bet and how skillfully they played, a casino could create digital personas that could then be used to adjust its marketing efforts, such as offering free play and giveaways to loyal customers.

    Individual chip and card attributes can be analyzed using computer vision.

    Insights from chips and cards are combined to digitize game play and understand the strategy behind the blackjack hand.

  2. Scene variety: Computer vision technology also can be used to visually target customers standing in line. When a customer queues up to buy coffee in a popular café, rent a car or check out in a convenience store, the most frustrating part is often long wait times. However, computer vision helps stores detect the length of the queue by counting the number of people that enter and leave it. Stores can then optimize the number of service reps based on demand and deliver a better customer experience. In the future, machines could also analyze the emotions of people in queues to assess whether they need more attention from employees. The possibilities are endless.

    Scene variety is also used to help marketers create more efficient store shelf positioning for products. Brand teams generally have contracts with retailers with specific instructions on product display. To ensure compliance, brand teams engage in costly monitoring by having sales reps visit stores and inspect the displays to make sure the products adhere to proper shelf placement, sequence, display price, varieties in stock and floor positioning, but marketers should consider using machines to speed up and simplify this process, too. Instead of conducting manual assessments, reps can take photos of the displays with their cell phone cameras and use machine systems to complete faster automated comparisons of shelving and stocking requirements. These vision-powered systems help the brand get maximum customer attention.

While computer vision is nothing new, it’s opened up many possibilities for companies across industries, and it’s here to stay. Are you ready? Thinking about the business problems that could be solved by computer vision is the first step to driving more effective marketing efforts and higher profits for your company.

Download this article for insights on how AI is enhancing the customer experience, and how your firm can take part.

Contact Us

We’d love to hear from you. Send us a message and we’ll respond as soon as possible.


Rasvan Dirlea

Rasvan Dirlea is a principal and leader of ZS’s pricing and revenue management practice for travel and transportation.

email linkedin
Naman Gandhi

Naman Gandhi is a consultant in ZS’s advanced data science space, where he focuses on machine learning.

email linkedin
Nilesh Kadam

Nilesh Kadam leads ZS’s India-based data science team, specializing in artificial intelligence and machine learning.

email linkedin