Golf swing analysis now runs on deep learning models that track 30 body points
Google Cloud built a system that uses deep learning with proprietary 2D and 3D models. The model tracks more than 30 key points on the golfer, club, and ball. It records top of swing, impact, follow-through, and finish for each shot.
Athletes move from video review to quantitative pose data. Coaches replace subjective feedback with exact joint angles and timing. This shifts training from feel-based adjustments to measurable corrections.
Bryson DeChambeau partnered with Google Cloud to test the system on his swing. The data let him adjust club path and body rotation with millimeter-level precision.
Step 1: Open Google Cloud Vertex AI at https://cloud.google.com/vertex-ai and create a new project. Step 2: Upload swing video to the Pose Estimation model and select the golf preset. Step 3: Review the 30-point skeleton output and export joint angle CSV for your next session.