Scalable Injury-Risk Screening in Baseball Pitching From Broadcast Video

2026-03-05

AI Summary

This study addresses the limited scalability of current pitching injury risk screening, which relies on costly multi-camera systems and is thus impractical in non-professional settings. The authors propose a method to reconstruct 18 clinically relevant biomechanical metrics from monocular broadcast videos by integrating pose estimation with global trajectory recovery. Key innovations include a drift-controlled global lifting module and a kinematic optimization pipeline that incorporates skeletal constraints, joint limits, and bilateral symmetry. Building upon DreamPose3D, the approach further introduces velocity parameterization, sliding-window inference, inverse kinematics, and temporal smoothing. Evaluated on 13 professional pitchers, the system achieves mean absolute errors below 1 degree for 16 of the 18 metrics. In a large cohort of 7,348 pitchers, it yields AUCs of 0.811 and 0.825 for predicting Tommy John surgery and major arm injuries, respectively.

Technology Category

• Computer Vision: Biometrics, Face, Gesture & Pose
• Intelligent Robots: State Estimation
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Security and Privacy: Large-scale security measurements

Abstract

Injury prediction in pitching depends on precise biomechanical signals, yet gold-standard measurements come from expensive, stadium-installed multi-camera systems that are unavailable outside professional venues. We present a monocular video pipeline that recovers 18 clinically relevant biomechanics metrics from broadcast footage, positioning pose-derived kinematics as a scalable source for injury-risk modeling. Built on DreamPose3D, our approach introduces a drift-controlled global lifting module that recovers pelvis trajectory via velocity-based parameterization and sliding-window inference, lifting pelvis-rooted poses into global space. To address motion blur, compression artifacts, and extreme pitching poses, we incorporate a kinematics refinement pipeline with bone-length constraints, joint-limited inverse kinematics, smoothing, and symmetry constraints to ensure temporally stable and physically plausible kinematics. On 13 professional pitchers (156 paired pitches), 16/18 metrics achieve sub-degree agreement (MAE $<1^{\circ}$). Using these metrics for injury prediction, an automated screening model achieves AUC 0.811 for Tommy John surgery and 0.825 for significant arm injuries on 7,348 pitchers. The resulting pose-derived metrics support scalable injury-risk screening, establishing monocular broadcast video as a viable alternative to stadium-scale motion capture for biomechanics.

Problem

• injury prediction
• biomechanics
• baseball pitching
• scalable screening
• monocular video

Innovation

• monocular video
• biomechanics recovery
• drift-controlled global lifting
• kinematics refinement
• injury-risk screening