A CLOSER LOOK AT YOUR SET
Review your deadlift,
rep by rep.
Check the plate path. Correct the tracking.
Compare two reps
from the same set.
Conventional deadlift · Fixed side view
Prototype input: H.264
MP4, 30/60 fps, up to 45s and 100 MiB.
Experimental beta: tracking accuracy and phone compatibility are not yet validated. Check every path against the video.
Browser processing · No account · No video upload
Aligned at lift-off. Look closer before drawing conclusions.
Original video only
Timing and trajectory results are unavailable for this file. If your browser can play it, you can inspect the original video below.
YOUR LOCAL REVIEW
Start with the frame.
Use a clear frame, select the camera-side plate center, and fit the reference circle. Arrow keys on the video move the point; Shift + arrows step frames.
THE SAME SET, SIDE BY SIDE
See where they differ.
Aligned at lift-off · Real elapsed time · Shorter side holds at its end
Plate path
Plate-diameter units · Both paths share the same scale
Relative
2D displacement, not centimetres or a technique score.
Measurement details & path data
| Rep | Time (s) | X | Y | Source |
|---|
Keep the observation.
A PNG review card, generated in your browser.
SMALL STEPS. A CLEARER VIEW.
A replay you can question.
Choose your clip
Use one conventional deadlift set filmed from a fixed side view. No account needed.
Check the path
Select the visible plate. Review tracking and correct a point when it loses the target.
Compare the reps
Align at lift-off, watch the original movement and keep a concise observation.
A USEFUL CLIP STARTS HERE
Give the plate a clear view.
- Keep the camera fixed, level and side-on.
- Keep the same camera-side plate visible throughout.
- Include lift-off and the end of upward movement.
- Use normal-speed H.264 MP4 at 30 or 60 fps.
This prototype cannot reliably detect every camera angle, occlusion or slow-motion edit. It does not evaluate injury risk or prescribe training.
Phone compatibility and tracking accuracy are still being evaluated. The motion test uses generated shapes, not real training footage.