Technical methodology
How VertChecker measures a vertical jump
VertChecker estimates jump height from the time between takeoff and landing. Computer vision proposes those events; projectile physics converts the reviewed flight time into height.
1. Decode the original timing
The analyzer reads video timestamps and nominal frame rate instead of assuming every file is constant-frame-rate. It rejects or warns on clips whose playback timing appears stretched, truncated, or otherwise inconsistent with normal ballistic motion.
2. Track the athlete and floor
A pose model follows body landmarks through the clip. Foot landmarks are preferred for ground contact; an ankle-based fallback is labeled when detailed foot landmarks are unavailable. Background motion can be estimated to stabilize modest camera movement, but a fixed camera remains the better protocol.
3. Propose takeoff, apex, and landing
Contact and trajectory signals generate candidate events. The system checks that the airborne path resembles a physically plausible parabola and surfaces quality warnings when the track, camera, floor, or landing technique weakens the estimate.
4. Convert flight time to height
h = g × t² ÷ 8
Here t is the elapsed time from takeoff to landing and g is 9.80665 m/s². This is a flight-time estimate of centre-of-mass rise. It is not a direct measurement of hand reach or the distance between a standing mark and a jump touch.
5. Report uncertainty and allow correction
A frame interval is 33.3 ms at 30 fps, 16.7 ms at 60 fps, 8.3 ms at 120 fps, and 4.2 ms at 240 fps. That timing resolution is not the same as end-to-end accuracy: contact visibility and jump technique can shift the selected events by multiple frames. VertChecker therefore shows event frames, warnings, and a frame seeker instead of presenting automatic timing as ground truth.
Validation status
Published studies support the underlying flight-time method and show that carefully selected high-frame-rate video events can agree closely with force platforms. Those studies do not validate VertChecker's automatic detector. VertChecker has internal physics and regression tests, but it has not yet published an independent force-platform validation study. Do not use it for clinical decisions, injury clearance, or claims that require certified equipment.
References
- Balsalobre-Fernández et al. (2015), validation and reliability of video-based vertical-jump timing against a force platform.
- Haynes et al. (2019), validation of manual 240 fps video event selection for drop-jump RSI.
- Statistics Canada (2019), age- and sex-specific jumping-height percentiles measured with a force plate.