How Facial Recognition Software Can Improve Your Event
There’s a scene in “Minority Report” (a film set in the not-too-distant future) where a character is trying to stay incognito, yet every advertisement he passes welcomes him by name. He even walks into a Gap only to have a hologram ask, “How’d those assorted tank tops work out for you?”
Let’s just table the fact that this guy is buying assorted tank tops, to acknowledge that, what seemed fantastical in 2002, is just around the corner in 2019 thanks to facial recognition software. In fact, facial recognition became such an everyday part of our lives so fast we sort of missed it happening. Currently, social media uses it to tag photos, airports (and some airlines) use it to screen passengers, and Apple uses so you can access your phone.
The software is not “Minority Report”-level omnipresent…yet. Retailers are not greeting you by name thanks to the software, but your event staff could. Facial recognition software may be on the verge of revolutionizing the events industry.
How Facial Recognition Software Works
When you spot someone you know in a crowd, what is it that clues you to their identity? It’s a combination of factors: the curve of their cheek, the bridge of their nose, the shape of their eyes, etc.
Facial recognition software works similarly by building an algorithm from the unique features of a person’s face. It does this through a technique known as point matching, where geometric landmarks are identified and mapped. So, the distance from forehead to chin, the space between eyes, and much more, all become part of an exclusive “faceprint.”
When facial recognition software scans you in a crowd, that faceprint, which is actually a mathematical formula, is matched to a database containing more than a hundred million identified faces. As long as your face is in that database, there’s a high likelihood that the software will identify you.
Although, according to an article in The Guardian, the software is not perfect. Even the best systems have trouble distinguishing between twins. There is also a troubling issue of racial bias that programmers need to address. “If a system is trained on a million white male faces, but fewer women and people of color, it will be less accurate for the latter groups. Less accuracy means more misidentifications.”
However, the article goes on to point out that “independent tests by the US National Institute of Standards and Technology (NIST) found that between 2014 and 2018 facial recognition systems got 20 times better at finding a match in a database of 12m portrait photos. The failure rate fell from 4 percent to 0.2 percent over the period, a massive gain in accuracy.”
Suggestion for Using Facial Recognition Software at an Event
At an event, the check-in process is often an attendee’s longest wait. It is also, frequently, the time when their willingness to wait is at its lowest. At the start of an event, most attendees want to quickly move through the check-in process so they can get started.
This is where facial recognition can shine. If you ask attendees to opt-in when they register, they can load a picture to your facial recognition system using the camera on their smartphone or webcam. Then when they arrive, all they need to do is look in a camera, and the check-in process is practically over before it began. Another helpful feature is that these stations do not necessarily need staff. A kiosk with a camera (or even a laptop on a table) is all that’s required.
Studies have shown that using facial recognition makes the check-in process five times faster than traditional methods.
Facial recognition software acts as an added layer of security when checking ID cards. The software can help to ensure that the person presenting the ID matches the image on the card. This process can also occur at registration or any time before the event begins. Ask attendees to scan their ID (such as a driver’s license, passport, etc.) and upload a recent photo.
Facial recognition software can also identify people who snuck into your event without checking-in and notify security personnel. While security is on their way, the software can also discover if the person has a criminal record and predict if they pose a threat. In fact, you can utilize known watchlists, so the software can spot if any troublemakers appear and notify onsite security immediately.
Currently, to identify a visitor, booth staff needs to manually scan a badge to ensure that all necessary information makes it into their lead management system. With facial recognition, that information is immediately added as a person approaches the booth. Additionally, booth staff can now greet each visitor by name. This added level of personalization will drastically improve a visitor’s experience.
The software also makes it easy to track how many people visited a booth and how long each one stayed. Advanced systems can even analyze how visitors were feeling throughout their stay based on facial expressions. At the end of the day, the software can deliver a report listing each visitor and the likelihood they will do business with a brand. This way, the sales team can reach out to the most promising leads with phone calls and in-person visits while still maintaining contact with the rest via email and social messaging.
A high level of personalization does not have to stop at the booth. Event staff, such as those maintaining the information area, can recognize visitors as they approach and possibly anticipate their needs. This is also true for food and beverage service personnel. In fact, it may be possible to pull a “Minority Report” and ask how someone enjoyed their caramel macchiato yesterday.
With a credit card on file, it is possible to use facial recognition for cashless payments. So, not only could a barista ask about yesterday’s order, they could make a new one and have the visitor on their way before they had time to stop at the counter. If they guess wrong, no big deal – it’s $.50 of coffee down the drain, and a fun story for the attendee.
Facial recognition can also be used in conjunction with chatbots to provide unparalleled personalization right on someone’s phone, such as real-time recommendations for sessions or activities.
Traffic Monitoring and Insights
Organizing pedestrian flow is a science. Often, your layout and planning work can be near perfection; other times, not so much. Maybe an attraction draws larger crowds than anticipated or the taco vendor proves to be immensely popular. Whatever the reason, lines have formed, pedestrians are impeded, and everyone is grumpy about it.
Utilizing facial recognition software can provide you with a real-time, bird’s-eye-view of your event. You can monitor as bottlenecks begin to form and quickly deploy staff to manage the situation and direct traffic. However, you can also delve deeper and examine where specific demographics clustered and areas they avoided. You could even follow a single attendee to gain insights on their entire event journey. The level of insights you wish to achieve is entirely up to you.
You’ll also be able to gain a better understanding of your event’s flow, where people tend to gather, and areas that are unexpectedly barren. This will help you better understand what is working and where some problems may exist. By sharing this information with vendors and exhibitors, you may be able to help some of the slower areas generate more foot traffic. If you discover there are certain areas where attendees naturally gather, you can charge a premium for those slots at upcoming events.
By leveraging the wealth of possibilities provided by facial recognition software, you can deliver a more personalized and memorable experience for your attendees, exhibitors, and sponsors.