Observed arrival · 2026-09-15
PongAI Reads a Table-Tennis Match from Body Movement
An AI analysis tool that turns side-on table-tennis video into shot timing, player attribution, stroke classifications, and 13 body-based measurements.
Field notes
The analysis pipeline uses body pose rather than the ball, which defines both its practical scope and its exclusions. The public example covers a three-minute match, reports 40 shots and 10 rallies at 120fps, and displays player attribution alongside the shot timeline. The page says the system can derive 13 measurements, including trunk rotation and recovery time, but deliberately leaves spin, racket angle, ball placement, and bounce timing unmeasured.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue