Can a smartphone camera measure jump height?

Vertical jump height is usually measured with equipment most people will never have access to — force plates, motion-capture rigs, or dedicated jump mats. But the physics behind the measurement is simple enough that it's worth asking whether a phone camera can approximate it.

The underlying method is flight-time kinematics: track the moment a person's feet leave the ground and the moment they land, compute the time spent airborne, and convert that directly into a height estimate using standard projectile motion equations. No depth sensor required — just an accurate estimate of two timestamps.

The catch is that 'accurate' is doing a lot of work in that sentence. A modern pose-estimation model can track hip and ankle keypoints frame by frame, but the keypoints are noisiest exactly at takeoff and landing — the two moments the entire calculation depends on. A one-frame error at 30fps is roughly a 33-millisecond error, and at typical jump heights, that's enough to visibly shift the final estimate.

That reframes the engineering problem: it's not really about pose estimation accuracy in general, it's about minimizing error at two specific, adversarial frames. Frame rate turns out to matter more than most of the other variables tested — higher frame rates shrink the timing granularity of the error directly. Camera angle matters too, but less obviously: tolerance for off-axis capture is narrower than intuition suggests, because perspective distortion changes the apparent vertical displacement of the tracked keypoints.

So: can a smartphone camera measure jump height? Provisionally, yes — closely enough to be useful for tracking relative progress over time, if not yet a substitute for lab equipment in absolute terms. The more interesting result is how much the answer depends on treating takeoff and landing as a distinct, higher-precision sub-problem rather than just another part of the trajectory.