Study at a glance

In 13 healthy adults performing a 1-minute sit-to-stand test, Theia3D showed strong to excellent agreement with marker-based motion capture for the dominant whole-body movement patterns underlying the task, which together explained over 98% of total movement variance, while agreement was weaker for smaller, higher-order coordination patterns, particularly those involving pelvic motion.

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: Theia3D version v2023.1.0.3161, patch 14, as stated in the paper.

Study Overview

Design

A concurrent validity study comparing whole-body movement synergies, extracted using principal component analysis, between a marker-based and a markerless (Theia3D) motion capture system during the 1-Minute Sit-to-Stand Test performed under eyes-open and eyes-closed conditions.

Participants

Thirteen healthy adults (7 women, 6 men; mean age 30.5 years; mean height 171.2 cm; mean weight 66.1 kg; mean BMI 22.4), who self-reported no musculoskeletal, neurological, or vestibular impairments within the previous six months.

Data Collection

Participants wore tight-fitting athletic sportswear and performed trials barefoot. Each participant completed two randomly ordered 1-Minute Sit-to-Stand trials, once with eyes open and once with eyes closed, separated by a 3-minute rest, rising from a standardized armless chair as fast as possible for 60 seconds with their feet shoulder-width apart. A 10-camera marker-based system (39-marker Plug-in Gait marker set) and an 8-camera markerless system processed through Theia3D recorded each trial simultaneously at 200 Hz, calibrated within a shared laboratory coordinate system and temporally synchronized via a shared hardware clock.

Analysis

Principal component analysis was performed separately on each system's own kinematic data, using marker trajectories for the marker-based system and model-derived joint center positions for the markerless system, to preserve each system's native coordinate structure. This produced a set of "principal movements," ranked by how much of each participant's total movement variability they explained, along with their corresponding position, velocity, and acceleration waveforms over time. Waveform similarity between the two systems' corresponding principal movements was quantified using Pearson correlation coefficients, interpreted as weak (below 0.50), moderate (0.50-0.75), strong (0.75-0.90), or excellent (above 0.90). The effect of removing visual input on the relative variance explained by each principal movement was additionally compared within each system using paired statistical tests.

Key Findings

  • The three dominant movement synergies, representing the rising phase itself, trunk flexion, and backward hip translation, were nearly identical between the two systems and together explained more than 98% of total movement variance in both (cumulative variance explained: 99.49% for Theia3D; 98.55% for the marker-based system), with very strong to excellent waveform correlations for these three dominant components (position correlations of 0.99, 0.96, and 0.89, respectively).
  • A specific higher-order movement pattern involving horizontal hip adduction-abduction coupled with vertical head oscillation was also identified consistently by both systems, with a correlation exceeding the strong-to-excellent threshold (r = 0.85), showing that at least some smaller-magnitude coordination patterns, not just the three dominant ones, were captured comparably by both systems.
  • Agreement was considerably weaker for most other higher-order movement patterns, which individually explained less than 1% of total movement variance each, with several of these patterns identified by each system at a different relative order of importance rather than at the same rank, a phenomenon the authors note is a known characteristic of independently computed principal component analyses rather than necessarily reflecting a genuine difference in underlying coordination strategy.
  • Both systems detected a measurable shift in coordination strategy when visual input was removed, consistent with reduced reliance on the dominant rising-phase movement pattern and increased reliance on secondary coordination patterns, but the two systems flagged a different, only partially overlapping set of specific movement patterns as significantly affected by the loss of vision. The marker-based system detected changes concentrated at both the dominant pattern and several higher-order patterns, while Theia3D primarily detected changes in patterns associated with trunk flexion and chest motion.
  • Theia3D did not capture pelvic movement patterns with the same fidelity as the marker-based system, a limitation the authors note is consistent with previously reported systematic offsets in pelvic tilt and weaker sagittal-plane correlation for Theia3D in other populations, suggesting pelvis-specific markerless measurement may warrant particular caution in this and other types of functional movement analysis.

What This Means for Movement Science Researchers and Functional Assessment Teams

For researchers and teams using the Sit-to-Stand Test or similar repetitive functional assessments, this study offers an encouraging, specific finding: Theia3D reliably captured the dominant whole-body coordination patterns, rising, trunk flexion, and hip translation, that together account for the overwhelming majority of movement variability in this task, supporting its practical use for functional screening, longitudinal monitoring, and assessment settings where a full marker-based laboratory setup is impractical. At the same time, the weaker agreement observed for smaller, higher-order coordination patterns, and for pelvic motion specifically, is a useful, concrete caution: subtle compensatory strategies, such as small trunk rotations or chest-hip coupling patterns that might indicate early functional decline, should currently be interpreted as exploratory rather than definitive when derived from markerless data, until further validation establishes how reliably these smaller-magnitude patterns can be captured.

The Role of Theia3D in Movement Analysis

Theia3D was the markerless motion capture system evaluated in this study, compared directly against a marker-based reference system not at the level of individual joint angles, but at the level of whole-body movement synergies extracted through principal component analysis during a repetitive, dynamically demanding functional task. This synergy-based validation approach offers a different, complementary kind of evidence to joint-angle comparisons, testing whether Theia3D captures the same underlying coordination structure a person actually uses to organize a complex movement, rather than only whether individual joint angles at each instant align with a marker-based reference.

Study Limitations and Future Directions

The authors note that agreement was reduced for higher-order movement synergies, which individually explained less than 1% of total movement variance and are inherently more sensitive to measurement noise and to modeling differences between systems; while less relevant to understanding gross functional performance, they note these smaller components may still provide insight into subtle compensatory strategies worth further investigation. The study sample was limited to healthy young adults performing a single functional task, which limits how well these findings generalize to older adults, people with movement-related differences, or other functional and movement-recovery-relevant tasks. The authors identify extending this kind of synergy-based markerless validation to gait, balance perturbation tasks, and other movement-recovery-relevant assessments, as well as to broader and more diverse populations, as important next steps before markerless synergy analysis can be more confidently applied in those contexts.

Citation: Promsri, A., Wargel, A., & Federolf, P. (2026). Concurrent validity of Theia3D markerless compared to marker-based motion capture for movement synergy analysis during the 1-minute sit-to-stand test. Journal of Biomechanics, 202, 113293. https://doi.org/10.1016/j.jbiomech.2026.113293

Link: Read the full study

Does Markerless Capture Reproduce the Same Movement Coordination Patterns as Marker-Based Systems?

This study goes beyond individual joint angles to show that Theia3D captures the same dominant whole-body coordination patterns as marker-based motion capture during a repetitive sit-to-stand task, accounting for over 98% of total movement variance, while agreement was weaker for smaller, higher-order patterns involving pelvic motion. If your work involves functional movement screening or coordination-based assessment, talk to our team about applying Theia3D to your own movement synergy research.

Recent Posts