Golf Swing Analysis Now Uses 30-Point Skeletal Tracking
Bryson DeChambeau's team applies Google's deep learning system with proprietary 2D and 3D models that track over 30 body, club, and ball points. The model records swing phases including top of swing, impact, follow-through, and finish to measure shot outcomes.
This replaces subjective coaching notes with precise kinematic data. Athletes gain objective metrics for every joint angle and club position rather than relying on memory or video replay alone.
Bryson DeChambeau partnered with Google Cloud to run the system during practice sessions, generating frame-by-frame swing data that directly informs equipment adjustments and training drills.
Step 1: Open Google Cloud Vertex AI at https://cloud.google.com/vertex-ai and create a new project. Step 2: Upload swing video files and select the Pose Detection model to extract 30 skeletal keypoints per frame. Step 3: Export the JSON output containing joint angles and club positions for each swing phase.