For over a century, tracking shots have been the bane of many a camera operator’s existence. Trying to keep your subject in frame and in focus can be tough no matter the conditions, but throw in crowds, weather, and fast-moving protagonists, and you quickly realize why NFL sideline cam ops get paid the big bucks. But a relatively new technology—automated subject tracking—has been making major waves in the video community, by doing away with those long-standing headaches. Suddenly, even relatively inexpensive PTZ cameras are capable of locking onto a subject and following them around complex shooting environments. How can you take advantage of these breakthroughs? Read on!
It's hard to find details on exactly when auto tracking started showing up on PTZ cameras, but it’s safe to say the technology has really hit its stride in the past few years. Especially with the rise of artificial intelligence-fueled machine learning, cameras have become extremely good at differentiating subjects within a frame. For a long time, automated tracking would falter when subjects moved too quickly, turned away from the camera, passed by other people, or didn’t have enough differentiation from the background. Now, many of these issues have been significantly reduced, if not eliminated entirely, and PTZ cameras can easily follow people in a classroom, office, auditorium, house of worship, and other similar environments. Tracking athletes on a field is still one of the trickiest tasks, but higher-end PTZs are up for the challenge, especially when they can use a mix of software and hardware tracking.
Many lower-cost PTZ cameras solely rely on software-based tracking. The video feed is sent into a computer, where software in the cloud or installed on the hard drive will analyze the feed, pick out the subject, and send commands back to the camera so that the subject stays in frame. This can be adequate for many situations, but isn’t quite as instantaneous as camera-based processing, and can become extremely taxing on the computer being used, especially if it’s guiding tracking for multiple cameras.
Hardware-based in-camera tracking typically works off a processor chip in the camera unit, with no need for any separate computing power. This is more streamlined, but often lacks the latest AI-based advancements and machine learning algorithms that get smarter over time. Hybrid auto-tracking cameras use hardware processing as a base, but also interface with software to deeper analyze the data and improve in real time.
There are more options within auto tracking—blocking zones, subject identification, zoom presets—but those will be heavily dependent on the make and model of your camera and corresponding software, so those options are best learned about through an accompanying manual.
What can auto tracking do for you? By eliminating the need for a dedicated and skilled camera operator, you’re suddenly able to get dynamic tracking shots at the push of a button. The most popular use case is educational settings, where classroom PTZ cameras can track teachers and presenters to aid in remote learning and lesson archiving. Having a static camera angle isn’t just boring, it can present major problems if subjects move out of their primary position. Auto-tracking PTZs can easily follow them around and provide a more intimate, immersive experience for remote learners. Houses of worship have embraced auto-tracking PTZs for similar reasons, often lacking staff or volunteers that can smoothly operate a tracking shot from the back of the room. Having a moving camera for a service can dramatically improve enjoyment and engagement for remote viewers.
Auto tracking is popular in corporate, conference, and event spaces, but does come with some caveats. There will certainly be times when you want to stream or record a presentation but don’t have a camera operator on hand. This is where auto-tracking cameras come to the rescue, adding a much more interesting feel to your footage than the classic “back of room static shot.” It’s important to note though, even the most advanced tracking software doesn’t have the knowledge base and thinking-on-your-feet skills of an experienced camera operator. For larger, higher-budget events where the video feed might be streaming out in real time and displaying on screens in the room, mistakes aren’t an option and auto-tracking usually won’t cut it—you need an operator that can respond to changing conditions rapidly and intelligently. Similar plusses and minuses apply in sports broadcasting. Auto-tracking cameras can save you time, money, and space, but fast moving subjects are hard, athletes in matching uniforms are typically passing by each other in a frame, and automated PTZs are liable to make mistakes. If you’re streaming out a middle-school basketball game, that might be a risk you’re willing to take (if the auto tracking loses its subject, you can step in and reset things), but if you’re doing video for Madison Square Garden, you probably want to go with a human operator.
Auto tracking in PTZ cameras will continue to evolve, and the rise of AI-based learning will make them smarter, faster, and more capable than ever before. For now, they lack the natural instincts and problem solving skills of a cameraperson who’s been in the business for decades, but for lower budget productions and situations where there wouldn’t be money for these capable camerapeople anyway, an auto-tracking PTZ is certainly superior to your iPhone propped against a book in the back of the room.
What do you use auto-tracking PTZs for? Let us know in the comments section, below. And if you have questions about PTZ cameras, feel free to give us a call, start a chat, or come visit us at the NYC SuperStore.
