Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Scalable Injury-Risk Screening in Baseball Pitching From Broadcast Video

Jerrin Bright, Justin Mende, John Zelek

TL;DR

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 is presented, establishing monocular broadcast video as a viable alternative to stadium-scale motion capture for biomechanics.

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.

Scalable Injury-Risk Screening in Baseball Pitching From Broadcast Video

TL;DR

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 is presented, establishing monocular broadcast video as a viable alternative to stadium-scale motion capture for biomechanics.

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 ). 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.
Paper Structure (29 sections, 5 equations, 2 figures, 2 tables)

This paper contains 29 sections, 5 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Sports Analytics Research.
  4. Biomechanical Analysis.
  5. Injury Prevention and Prediction.
  6. Methodology
  7. Global normalization.
  8. Pelvis-to-Global Lifting Module (PGLM)
  9. Biomechanics Refinement Stack (BRS)
  10. Bone-length enforcement.
  11. IK fit with joint limits.
  12. Symmetry constraints.
  13. Handedness Estimation
  14. Biomechanical Metric Computation
  15. Temporal Derivatives
  16. ...and 14 more sections

Figures (2)

  • Figure 1: Injury-Risk Screening Pipeline. Monocular video yields 3D pose, which is lifted to global space and refined with kinematic constraints to produce biomechanical metrics; these metrics are aggregated into clinically motivated features for injury-risk prediction.
  • Figure 2: Feature Category Importance for Injury Prediction. Distribution of GBM gain-based feature importance across nine feature categories for three injury targets