Computer Vision in Theme Parks: 9 Use Cases
When the Walt Disney Company unveiled the Magic Band in 2013, they struck a data gold mine. This convenient wearable tech allowed Disneyland attendees to purchase food, souvenirs, and enter rides, creating a much more convenient and streamlined park experience for visitors.
But under the hood, this little device was taking notes: the wristband was tracking the wearer’s movement and actions, while algorithms crunched away at the numbers to predict everything from their preferred eating times to what rides they’d go on next.
The data Disney collects is invaluable and has undoubtedly helped boost revenues. But there’s one problem: not everyone that attends a Disney amusement park opts to get a Magic Band. Despite the fairly low starting price of $19.99, that barrier to entry means that Magic Band data will be limited to only a subpopulation of visitors, which makes drawing generalized conclusions from its data more difficult.
For optimal data collection, theme parks need a technology that collects data universally for all attendees. Luckily, that technology already exists.
Computer vision leverages AI technology to draw insights from visual inputs. That means that your guests don’t need to wear a bracelet for you to collect data about their movements. Instead, you can do so the old fashioned way: with your computer’s own “eyes.”
As the theme park industry continues to recover from the COVID pandemic, this technology is more important than ever. Although theme park attendance is still down compared to the pre-COVID era, revenue is higher than before because guests are spending more per visit. Powerful AI technologies like computer vision can play an important role in continuing to drive this strong visitor spend.
What is Computer Vision?
Computer vision (CV) is a field of AI that uses visual inputs to collect and interpret data. In other words, it’s a discipline that focuses on trying to get computers to see the world the way we do — or even better than we do.
Chances are you’re already acquainted with CV to some extent: if you have an iPhone, you use it every time you unlock your phone with FaceID. If you’ve ever tried on a Snapchat filter, you’ve used it as well.
In commercial applications, however, CV is usually used to draw insights from that visual data. For example, restaurants may use CV to estimate wait times, construction companies can use it to keep workers safe, and retail establishments can use it to track hot spots in their stores.
To achieve these feats, CV models convert images, videos, and real-time streams from security cameras into ones and zeroes. Then, they use neural networks, deep learning, and machine learning to interpret the data.
The result is a very powerful and versatile AI technology. While many AI models require data to be in a very specific format, such as GPS coordinates from a wearable bracelet, CV can draw insights from one of the most organic sources of data there is: the entire visual world.
9 Use Cases for Computer Vision in Theme Parks
As a whole, CV is a nascent industry, and its use in theme parks is even newer. That means that there’s lots of room for exploration and discovery of new ways to integrate it into business operations. Here are nine ways that amusement parks are already using computer vision.
Staff Detection
The importance of sufficient staffing has been more evident than ever during the past few years of COVID. As labor shortages worked their way across the world, businesses suffered. One look at the rate of order errors in the restaurant industry (26%) makes the magnitude of the issue readily apparent.
Computer vision can keep track of staffing and prevent shortages by discerning the number of staff members at different areas of the theme park. Combined with occupancy, pedestrian, and vehicle detection, CV models can monitor for disproportionate visitor to staff ratios across the park and redirect staff to areas that need to be bolstered due to sudden increases in demand. Over time, the technology can also be used to detect areas that are chronically understaffed.
Staff Engagement Monitoring
To create a memorable experience for your guests, staff quality is just as important as staff quantity. Guests rely on staff to guide them around the park and, in some cases, to entertain them — Disney theme parks are famous for their friendly and engaging costumed staff.
With CV, theme parks can monitor how often staff members are engaging with visitors. If facial recognition technology is used as well, the CV model can identify the staff with the highest engagement rates and managers can use them to train other team members. Similarly, CV can pick out staff with low engagement rates for further training.
Throughput Monitoring
Good strategic decisions always depend on accurate data. For theme parks, throughput is one of the most important metrics: without knowing where the guests in your park are going, it’s impossible to make informed decisions about how to allocate resources to different rides and attractions.
Computer vision can continuously monitor the flow of people through different points in the theme park, including rides, to derive insights about park performance. Those throughput measures can then be compared to different historical periods, attractions, and even different parks. From there, managers can develop better and more realistic business strategies.
Live Wait (Queue) Time Estimates
Managing customer expectations is an important part of any business. The old adage of “underpromise, overdeliver” holds true: if the reality of your offerings don’t live up to your promises and your customer’s expectations, your customers will push back, and you’ll likely lose business.
In theme parks, wait times are the most obvious area where customer expectation management is needed. If wait times for rides are not accurately estimated, guests will quickly become frustrated and exit lines to go to other attractions. If estimates are consistently inaccurate, they may leave the park entirely.
Although wait times are inevitable, it’s important to ensure that your customers are getting an accurate idea of how long they’ll be waiting. Computer vision can automatically calculate how long the average visitor will need to wait for a ride in order to provide more accurate wait time estimates, which will manage their expectations and mitigate line attrition.
Occupancy Monitoring
Theme park operators keep track of occupancy and availability metrics to determine how efficiently rides are being filled. Using real-time video, computer vision models can detect how many seats are available on different rides and make efforts to redirect traffic flow away for over-occupied rides and towards under-occupied ones. Disney, for example, uses Magic Band data to send personalized promotions through the My Disney Experience app to guests in order to influence where they go in the park.
Heat Map and Guest Journey Tracing
Using feeds from security cameras and Internet of Things (IoT) sensors, computer vision can keep track of where guests move in your theme park. Based on that data, CV models can create accurate heat maps and guest journey tracings that can shed light on where your guests are naturally gravitating to and what assets they’re using.
These maps can be used not only to improve efficiency but also to manage crowd flow, identify bottlenecks, prevent dangerous situations that can lead to crowd crush and develop insights for what to invest more resources into.
Guest/Door Counts
Guest/door counts monitor how many people are coming in and out of your theme park. Although theme parks typically have turnstiles at all entries that capture similar data on guest entrances, CV may be able to provide additional insights, such as whether people are sneaking in and data on guest exits if systems aren’t already in place for guest checkouts.
In addition, guest/door counts can be used within the parks to better understand different retail locations’ guest volume at different times of the day, which areas are the busiest and which locations might be better used as something else.
Parking Management
Like all crowded attractions, theme parks can struggle with maintaining streamlined parking lots. With CV, parks can monitor lots to identify bottlenecks, popular spaces, or areas where visitors seem confused. Park operators may also be able to use this data to identify prime spots for advertisements and promotions.
Interactive Experiences
Computer vision tech isn’t just for the business side of theme parks — it can also be integrated into attractions to create guest experiences. For example, in 2021, Disney unveiled a new Spiderman attraction that uses computer vision to simulate webs shooting out of visitors’ wrists. These sorts of augmented reality (AR) experiences will likely become more common as theme parks try to get more involved in the metaverse.
Key Takeaways: Computer Vision in Theme Parks
WIth attendance numbers still falling behind pre-pandemic numbers, theme parks need to use every tool in their arsenal to ensure customer spending stays high. Computer vision models can help park operators identify new ways to leverage opportunities and mitigate inefficiencies so that theme parks can stay profitable, create unforgettable guest experiences and maintain customer satisfaction.
Learn more about occupancy monitoring.