Observed arrival · 2026-10-05
A Curveball, Under the Microscope
An interactive physics lab simulates spinning-ball trajectories and uses AI to estimate motion parameters from a path or uploaded CSV.
- For
- Students and demonstrators exploring ball flight
- Worth noticing
- Uploaded CSVs use time_s,x_m,y_m; pixel coordinates must first be calibrated to metres.
Field notes
Users can switch between 2D and 3D observation, select tennis, soccer, table-tennis or smooth reference balls, and vary spin, launch conditions and wind. The inference workflow accepts time-and-position CSV data, shifts its first point to the origin, truncates at the first return to y≤0, and fits a 2D dynamics model; pixel coordinates need prior calibration to metres. The page says its smooth-ball drag relation is generic and that presets need calibration for surface texture and orientation.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue