Test frontal arm geometry.
Keep the full motion visible.
A marker-assisted frontal capture with three extended holds and two controlled bend-up/extend cycles.
Set up before recording
Fix the phone physically in landscape orientation, use the rear camera, do not zoom, and do not move the phone. Frame the crown, feet, full body, visible floor, full right arm, and all three markers. Face the camera; keep the right palm and elbow crease facing forward, and keep the elbow fixed during bending. Tap START DELAYED CAPTURE once, then move into position during the fixed ten-second delay.
The overlay is visual guidance only. Pose inference is disabled and no body geometry can disable, delay, cancel, or alter capture. Timed measurement commands use the unchanged decoded local audio files. No microphone, cloud inference, account, or required calibration object is used.
Audio / Protocol Timing Test
This runs the identical frontal goal-post clock and decoded local audio cues used by recorded capture. It never opens the camera or writes a recording. Use Camera + Audio Timing Test for realistic 4K recorder load.
Face the camera. Keep your full right arm and all three markers inside the frame.
A static scene is sufficient. Each supported format runs the same 4K recorder path, one-second pre-roll, 18.5-second protocol clock, timed audio, frame workload, and 750 ms post-DONE margin. Unsupported formats are skipped.
Real capture and Camera + Audio Timing Test share the same MediaRecorder, one-second pre-roll, frame workload, protocol dispatcher, and decoded audio path. Test-mode media is discarded.
Export sessions before clearing browser data. Aborted and interrupted attempts remain visible.
Manual research review
For frontal goal-post frames, manually verify the physical shoulder, elbow, and wrist marker identities. Legacy N0 sessions may still use crown/floor height anchors. Do not move marker points to make segment lengths look more consistent.
Per-block projected measurements
Spread between frames is not independent repeatability. Bent-pose shoulder-to-wrist chords are intentionally different from extended-pose chords. Compare like poses only.
Attempt-to-attempt manual baseline comparison
A change in projected length mixes annotation, pose, depth and camera errors. It is not a validated anatomical difference or a formal accuracy pass.