Study at a glance

In nine collegiate tennis players performing serves, Theia3D-based markerless motion capture closely tracked marker-based motion capture's joint angular velocity patterns at the knee, pelvis, trunk, and shoulder, with a consistent, predictable bias identified specifically in peak trunk rotation velocity.

Summary

Disclaimer: The following summary details independent academic research. Theia3D is intended for research, sports performance and educational use only. It is not cleared as a medical device and is not intended to be used for the diagnosis, treatment, mitigation or prevention of any disease, injury or medical condition.

Editor's note: This paper is a conference abstract rather than a full peer-reviewed journal article, presented at the American College of Sports Medicine's 2025 annual meeting. It does not report a specific Theia3D software version number, and, as is typical for abstracts of this length, does not include a full methods, results, or discussion section beyond what is summarized here.

Study Overview

Design

A concurrent comparison study evaluating agreement between marker-based and markerless (Theia3D) motion capture for joint angular velocity during the tennis serve, presented as a conference abstract.

Participants

Nine collegiate tennis players.

Data Collection

Participants performed five tennis serves each, with joint kinematics recorded simultaneously by a marker-based (Qualisys) system and a markerless system processed through Theia3D.

Analysis

Time-series angular velocity data for the left knee, right knee, pelvis, trunk, and shoulder were normalized from the trophy position to foot strike and compared between systems using Statistical Parametric Mapping (SPM), Concordance Correlation Coefficients (CCC), and Bland-Altman analysis.

Key Findings

  • Statistical parametric mapping identified significant differences in angular velocity between the two systems for all joints and segments studied, but only at specific phases of the movement, not throughout the entire serve.
  • When peak angular velocity specifically was examined, a significant difference between systems was found only for trunk rotation.
  • Peak trunk rotation velocity showed moderate agreement between systems (Concordance Correlation Coefficient of 0.539), with Bland-Altman analysis indicating a consistent average bias of about 100.7 degrees per second.
  • For all other joints and segments, and for peak values generally, statistical parametric mapping did not identify the differences between systems as significant, indicating that markerless motion capture closely tracked the same movement patterns as marker-based motion capture.
  • The authors conclude that markerless motion capture demonstrated high precision and accuracy in capturing both continuous joint velocity profiles and peak values during the serve, with the limited and consistent nature of the differences observed supporting its viability as a reliable, efficient tool for dynamic biomechanical assessment in tennis and similar sports.

What This Means for Movement Science Researchers and Tennis Performance Teams

For researchers and performance staff working with tennis players, this abstract offers an encouraging, sport-specific early finding: markerless motion capture tracked the overall angular velocity patterns of the serve closely across the knees, pelvis, trunk, and shoulder, with only one specific measure, peak trunk rotation velocity, showing a significant and consistent difference from marker-based motion capture. Because that difference was described by the authors as predictable and consistent rather than large or variable, it offers a specific, practically actionable note for teams using markerless capture for serve analysis: peak trunk rotation values may benefit from an adjustment to account for this known bias, while other measures tested can likely be interpreted with greater confidence as-is.

The Role of Theia3D in Movement Analysis

Theia3D was the markerless motion-capture system directly evaluated against marker-based motion capture in this study, specifically for its ability to capture the fast rotational demands of the tennis serve. The abstract's finding of generally strong agreement, with one specific, well-defined and consistent area of difference at peak trunk rotation velocity, offers a useful, sport-specific early data point for markerless capture's application to tennis and other rotationally demanding sports.

Study Limitations and Future Directions

As a conference abstract rather than a full published article, this summary is necessarily limited to the information the authors were able to include in a brief format: it does not provide full detail on participant characteristics, camera configuration, or discussion of potential sources of the trunk rotation bias identified. The sample size was small, consistent with many abstract-stage studies, and the authors do not report whether a full peer-reviewed manuscript expanding on these findings is planned or in progress. Readers interested in more detailed methodology, statistical reporting, or discussion of these findings should watch for a fuller published version of this work.

Citation: Valle, B. A., Snyman, K., & Aguinaldo, A. (2025). Concurrent validity and reliability of a markerless motion capture system in estimating joint kinematics in the tennis serve [Abstract]. Medicine & Science in Sports & Exercise, 57(10S), 52. https://doi.org/10.1249/01.mss.0001154940.13389.34

Link: Read the full abstract

Can Markerless Capture Track Tennis Serve Kinematics?

This study of nine collegiate tennis players finds that Theia3D markerless motion capture closely tracked marker-based motion capture's joint angular velocity patterns during the tennis serve, with a consistent, predictable bias identified specifically in peak trunk rotation velocity. If your work involves tennis or rotational sports performance research, talk to our team about applying Theia3D to your own studies.

Recent Posts