PoseTracker API logoPoseTracker

For product leaders

Range of Motion from the Phone Camera

Computerized range of motion from a phone camera: the joint angle a fitness or rehab app can show. Not a hardware goniometer or a diagnosis.

· Markdown

Computerized range of motion is a joint angle taken from the body in frame, not from a plastic goniometer. A fitness or rehab app needs that number when the protocol says “how far did this hold go?”. PoseTracker reads it from one phone camera: on-device MoveNet, 17 COCO points, the angle the picture actually contains.

People also search joint range of motion and range of motion measurement. The job is the same. The camera has to see the joint. A table of “normal” degrees does not see anyone.

What the camera can report

  • Flexibility holds the body shows face-on or at a clear angle: split, needle, pancake, bridge, and the other poses in the flexibility tool. Photo or live camera.
  • A side-view trunk fold with straight knees: trunk, hip and knee angles, plus a relative reach. That is the back flexibility test. It is not a tape on the skin and it does not print Modified Schober centimetres.
  • A custom hold you describe in the Assistant after sign-in, if the joints in the spec are among the 17 points and visible.

Keypoints stay on the device. The angle is the product output your UI shows.

What it cannot report

  • A hardware digital goniometer reading. Those results are devices you buy. This is a camera.
  • Normal clinical ranges (shoulder flexion “should be” N degrees, and the rest of that literature). PoseTracker does not publish them and does not decide if a shoulder is healthy.
  • A joint the phone cannot see: the far hip in a face-on shot, a rotation in the third dimension, a hand mesh.
  • Several people, a depth camera, or a 33-point BlazePose skeleton. The engine is MoveNet. If the protocol needs 3D or foot detail, start at MediaPipe vs MoveNet.

How a product uses it

Virtual physical therapy and coaching apps lose the session when the only artifact is a PDF. The virtual physical therapy page is that job. This page is the measurement: show the angle, store it, compare the next hold. The clinical meaning stays with the professional who wrote the protocol.

Ready-made path: open the flexibility tool. SDK path: the same MoveNet engine as the rest of PoseTracker, React Native, iOS, Android and web. Map of models and SDKs: what pose estimation is.

FAQ

What is computerized range of motion?

An angle computed from body keypoints instead of a plastic goniometer. PoseTracker reads 17 COCO points on the phone, on-device, and reports the angle the camera can see — for example a front split or a side-view trunk fold. It does not replace a clinician’s measurement.

Can a phone camera replace a digital goniometer?

No. Search results for digital goniometer are mostly handheld devices. A phone camera measures a joint that is visible in the frame, in 2D. It does not strap to the limb, and it does not output a clinical normal range.

Which ranges can PoseTracker show?

Holds the camera can see on 17 points: flexibility poses such as a split, needle, pancake or bridge in the flexibility tool, and a side-view trunk fold with straight knees in the back-flexibility test. A joint hidden by the body, or a 3D rotation the phone cannot see, is not a result.

Do you publish normal shoulder range of motion in degrees?

No. Normal ranges belong in a clinical reference, not in this SDK. PoseTracker returns the angle it observed. Interpreting that angle as healthy, limited or post-surgical is the clinician’s job.

Is this a medical device?

No. Movement analyses are pose-estimation opinions for the app that displays them. They are not a diagnosis, not a Modified Schober test, and not a substitute for an in-person assessment.

Need help building your fitness solution?

Want to learn how PoseTracker can enhance your fitness app? Contact us for a demo and see how easy it is to bring real-time tracking and personalized feedback to your clients.

BOOK A CONSULTATION

Or contact us :

— by email : fabrice@movelytics.fr

— by phone or WhatsApp : +33645200191

Try the Assistant · Fitness SDK guide · SDK overview