Study at a glance
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 systematic review rather than a single original study, so no individual Theia3D software version or camera hardware configuration applies. Theia3D is one of several markerless motion capture systems discussed across the 53 studies synthesized in this review (used in roughly 13% of the included studies, by the review's own technology breakdown), rather than a system the review authors themselves tested or validated directly.
Study Overview
Design
A systematic review, registered on PROSPERO and reported according to PRISMA 2020 guidelines, synthesizing the existing literature on the reliability and validity of markerless motion capture for measuring lower-extremity and trunk kinematics during sports-related and functional tasks (walking, squatting, jumping/landing, running, and cutting).
Literature Search
The authors searched PubMed, Web of Science, and CINAHL databases from the earliest available records through February 28, 2024, using terms related to markerless motion capture, validity/reliability, and the five target tasks, supplemented by manual reference-list screening. Of 409 records identified, 201 remained after duplicate removal; 82 articles advanced to full-text review following title/abstract screening, and 53 studies were ultimately included, 47 assessing validity against marker-based motion capture as the reference standard and 15 assessing reliability (including 9 studies that assessed both). Two authors independently screened studies for eligibility and performed quality appraisal using a standardized critical appraisal tool, with a third author resolving disagreements.
Content Reviewed
The review covers markerless motion capture technology characteristics across the included studies (camera type, number of devices, and software used for pose estimation and data processing); reliability findings broken down by task (walking, running, and squatting) and by within-session versus between-day measurement; validity findings broken down by all five tasks and by plane of motion; and a discussion of practical application to loading-risk-relevant movement screening, along with a structured risk-of-bias assessment of the included literature.
Key Findings
- Markerless motion capture's relative reliability ranged from low to excellent depending on the task, joint, and plane of motion, with most variables reporting measurement error (SEM) below 5°; squatting and landing tasks showed the highest reliability overall, with good-to-excellent agreement across most joints, the main exception being hip flexion, which showed measurement error ranging from 4.0° to 11.1°.
- Validity studies comparing markerless to marker-based motion capture showed considerable variability, with differences between the two technologies ranging from 0.2° to 28.6° depending on the task, plane of motion, and joint analyzed, and correlation strength ranging from negligible, including some negative correlations (particularly in the transverse plane), to very strong.
- The hip angle in the frontal plane reported the smallest differences between markerless and marker-based technologies across tasks, while walking showed the largest and most variable differences between technologies, likely reflecting the much larger number and diversity of studies examining this particular task.
- Depth-sensing cameras, including the Microsoft Kinect, were the most commonly used markerless technology (53% of included studies), though the review notes these cameras carry known practical limitations for sports biomechanics, including restricted capture range, lower frame rates, and difficulty operating in bright outdoor light; Theia3D was used in roughly 13% of the included studies, behind OpenPose-based pipelines (26%) and Microsoft's own development kit (19%) as data-processing approaches.
- The review's own conclusion is direct on the central validity question: while markerless motion capture provides reliable measurements for most kinematic variables studied, its results "are not largely comparable to those obtained using marker-based motion capture systems," underscoring that reliability and validity are related but distinct properties, and that good reliability does not by itself guarantee close agreement with marker-based measurement.
- The authors note that all 53 included studies involved only healthy individuals, and identify testing markerless motion capture's reliability and validity specifically in populations managing an existing lower-limb concern as an important and currently unaddressed direction for future research.
What This Means for Movement Science Researchers and Sports Performance Assessment Teams
For researchers and performance staff considering markerless motion capture for athlete movement screening, this review offers a specific, practically important synthesis: markerless systems generally provide reliable, repeatable measurements for most lower-extremity and trunk kinematic variables, particularly during squatting and landing tasks, but the degree of agreement with marker-based motion capture varies considerably by task, joint, and especially plane of motion, meaning sagittal-plane measurements can generally be trusted with more confidence than frontal- or transverse-plane measurements when direct comparability with marker-based benchmarks matters. The review's call for more standardized testing protocols, consistent reporting of practically meaningful statistics like measurement error and minimal detectable change, and broader task coverage (particularly for running, landing, and cutting, which remain comparatively understudied relative to walking) offers a concrete, specific roadmap for the kind of validation evidence the field still needs to build confidence in markerless motion capture for everyday performance assessment use.
The Role of Theia3D in Movement Analysis
Within this systematic review, Theia3D is discussed as one of several commercial and open-source markerless motion capture systems represented across the broader body of reliability and validity literature, appearing in roughly 13% of the 53 included studies spanning walking, running, and squat/landing tasks. The review's synthesis is evenhanded across systems rather than focused specifically on Theia3D, situating Theia3D-based studies within the wider landscape of markerless technologies (including depth-sensing cameras, single- and multi-camera video systems, and various pose-estimation software) that the field as a whole is still working to standardize and validate.
Study Limitations and Future Directions
The review's authors identify several limitations of the underlying literature it synthesizes. Considerable heterogeneity existed across the included studies in task characteristics (such as walking and running speed, and single- versus double-leg variations for squatting and landing), markerless motion capture systems used, and statistical analyses reported, which limited direct comparison between studies and required the authors to exclude some otherwise-relevant studies that used different statistical metrics than the majority. Certain tasks remain comparatively understudied: reliability evidence for landing and running is limited to only a handful of studies, and only a single study assessed the reliability or validity of cutting, meaning conclusions for these tasks are considerably less certain than for walking or squatting. Because gold-standard comparisons against biplanar videoradiography were not performed in any of the included studies, the authors note that the true accuracy of markerless motion capture, as opposed to its agreement with marker-based motion capture specifically, remains unknown. The authors also note a risk of bias across the literature, since none of the included studies reported whether raters were blinded to prior findings when conducting assessments, and most studies did not report examiners' qualifications or experience level, both of which could inflate apparent reliability. Looking forward, the authors call for standardized testing protocols, broader inclusion of loading-risk-relevant tasks matched to specific sport demands, consistent reporting of practically meaningful statistics such as measurement error and minimal detectable change, direct comparisons between different markerless systems to clarify whether any one system holds a meaningful advantage, validation against biplanar videoradiography as a true gold standard, and future research extending beyond healthy populations to those managing an existing lower-limb concern.
Citation: Yoma, M., Llurda-Almuzara, L., Herrington, L., & Jones, R. (2025). Reliability and validity of lower extremity and trunk kinematics measured with markerless motion capture during sports-related and functional tasks: a systematic review. Journal of Sports Sciences, 43(17), 1703-1730. https://doi.org/10.1080/02640414.2025.2518359
Link: Read the full study
How Reliable and Valid Is Markerless Motion Capture for Sports-Related Movement Screening?
This systematic review of 53 studies synthesizes the reliability and validity of markerless motion capture, including systems such as Theia3D, for measuring lower-extremity and trunk kinematics during walking, running, squatting, jumping/landing, and cutting, finding generally good reliability but variable agreement with marker-based motion capture depending on task, joint, and plane of motion. If your work involves sports performance screening or movement assessment research, talk to our team about applying Theia3D to your own studies.