Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Single RGB Camera Marker-less Motion Capture Systems (MLMCS) in Human Lower-Limb Activities: A Scoping Review
ABSTRACT
Background:
Single RGB-camera markerless motion capture (MLMC) is emerging as a low-cost, mobile alternative to laboratory-based, multi-camera or marker-based motion capture systems (MBMCS) for analysing lower-limb biomechanics. Yet, its research use, technical configurations, and validation practices have not been comprehensively mapped previously.
Objective:
To systematically map the extent, characteristics, and methodological approaches of single RGB-camera MLMC used to quantify lower-limb kinematics (hip, knee, ankle), by synthesising evidence on (i) participant populations and movement tasks, (ii) computational pipelines and human pose estimation algorithms, and (iii) reference standards and reported validity/reliability outcomes, in order to identify current evidence gaps and priorities for improving monocular accuracy and clinical applicability.
Methods:
Following JBI guidance and the PRISMA-ScR checklist, we searched PubMed, Scopus, IEEE Xplore, CINAHL, and SPORTDiscus (inception: February 22, 2025). Extracted data included participants’ characteristics, tasks, MLMC algorithms, hardware specifications, and validation approaches. Study quality was assessed using a modified 14-item Downs & Black checklist.
Results:
The five-database search yielded 3911 records; 28 met the inclusion criteria (published 2007-2025, with 86 % appearing since 2020). Studies spanned 17 countries and analysed 76 (± 190) participants (range 3-1026, mean age 29.6 (± 5.3) years). Gait, either walking or running, dominated the movement repertoire (23/28 studies, 82%), followed by single-event tasks (e.g., drop jump, hop). Twenty-three unique MLMC pipelines were identified (12 × 2D Model, 11 × 3D Model); OpenPose was the most frequent 2D method (10 studies). Where benchmarking occurred (25/28 studies), multi-camera MBMC was the predominant reference standard. Modified Downs & Black scores revealed a strong overall reporting, but significant limitations concerning external validity, blinding, and justification of statistical power.
Conclusions:
This review establishes that while single-camera MLMC systems are valid for sagittal plane hip and knee analysis, significant accuracy deficits persist for ankle and non-sagittal plane movements. The choice of the human pose estimation (HPE) algorithm is a paramount determinant of data quality and clinical viability.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.