Apple patents motion-predicting technology that can count reps and identify exercises during a workout stephen.warwick@futurenet.com (Stephen Warwick) | usagoldmines.com

• Apple has patented a new exercise-tracking prediction method
• It says the technology can capture body motion data and determine what exercise you are doing
• It could also count how many reps you’ve done, possibly paving the way for improved workout tracking or online class participation

While we’ve been hearing that Apple is preparing a pretty significant Health app upgrade for 2025, the company has also patented a new type of exercise tracking technology that could power improved workouts on many of the best iPhones in the future.

A new Apple patent published April 3 and seen by TechRadar, dubbed ‘exercise tracking prediction method,’ reveals Apple is exploring new technology for workout tracking. It pertains to “predicting and counting repetitions of physical activity” using captured image data of a body in motion.

Apple’s patent aims to overcome some of the limitations of predicting and tracking the activity of a person in real-time.

To that effect, Apple’s technology can take image capture data and predict what activity you’re performing while it is in progress and potentially provide feedback on it.

That could be done using a camera on any type of electronic device, including a phone, tablet, computer, or even a wearable like an Apple Watch (which is tipped to get a camera one day). There are heavy shades of Peloton Guide here, but it looks like Apple is trying to take the tech one step further with more precise tracking.

Apple exploring this technology could point to a number of implementations and upgrades for users, notably around using Apple devices to track workouts in real-time or when participating in a more collective version of an online Apple Fitness Plus class.

Apple explores workout detection

The best Apple Watches, Apple’s Health app, and devices like the iPhone excel in certain health and wellness tracking implementations; however, one obvious missing feature is driving workouts using rep counting and tracking.

Even the Amazfit Active 2, a $99 smartwatch we rate as one of the best cheap smartwatches on the market, has a built-in rep tracker for strength and conditioning workouts,

If Apple could crack workout and rep tracking, it would be a big boost to the company’s health and wellness ambitions, and using its device cameras could be a key to gaining the upper hand.

Powerful features like LiDAR would be much more accurate than accelerometers when it comes to tracking form and reps in a workout participant, giving Apple potentially unparalleled accuracy in this regard.

Given how generally cumbersome and often antisocial filming in the gym can be, it seems like this sort of technology would lend itself more readily to use in the home.

I can definitely imagine a service like Apple Fitness+ leaning on workout tracking data to give users more clinical insight into their progress, tracking reps on screen in real-time during a workout.

While not mentioned here, there’s also no reason to think this technology couldn’t be developed further to deliver feedback and pointers on form and posture, too.

At this point, the technology is only an idea on paper. If Apple ever releases a feature like this, it will likely be a year or two away from consumers. So don’t expect an Apple Watch Ultra 3 with a camera that can track your workouts.

More immediately, we know that Apple is planning a big blood pressure monitoring upgrade for the aforementioned Ultra 3 and the Apple Watch Series 11, both of which are expected later this year.

Latterly, rumors point to a new Health app overhaul, replete with an AI agent designed to replicate your Doctor, arriving as early as 2026.

You may also like

• An Apple a day? Your iPhone could soon have an AI Doctor thanks to a new iOS 19 Health app
• Using a smartwatch could be a game-changer for people with diabetes, new research suggests
• After 10 years of Apple Watch, the company reveals the one thing it got wrong about its users