AI models now dissect golf swings with 30 tracked body points
Google Cloud built a deep learning pipeline that records 2D and 3D coordinates for over 30 body, club, and ball landmarks. The system timestamps top of swing, impact, follow-through, and finish, then outputs shot outcome metrics.
Users stop relying on vague video review. They adopt quantitative feedback loops that replace opinion with measurable joint angles and club paths.
Bryson DeChambeau partnered with Google Cloud to test the pipeline on his own swing data and reported measurable changes in club-face angle consistency during practice rounds.
Step 1: Open Google Cloud Vertex AI and upload swing footage. Step 2: Enable the 2D/3D pose model at https://cloud.google.com/vertex-ai. Step 3: Review the generated CSV of 30-point coordinates and adjust grip or stance based on flagged angle deviations.