Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Monocular Markerless Motion Capture Enables Quantitative Assessment of Upper Extremity Reachable Workspace

Seth Donahue, J. D. Peiffer, R. Tyler Richardson, Yishan Zhong, Shaun Q. Y. Tan, Benoit Marteau, Stephanie R. Russo, May D. Wang, R. James Cotton, Ross Chafetz

TL;DR

This paper addresses the challenge of bringing objective upper-extremity mobility metrics into routine clinical care by validating a monocular MMC pipeline for UERW. The method combines PosePipe with MeTRAbs-ACAE for keypoint extraction, a MuJoCo-based full-body reconstruction via bilevel optimization, and a comparison of frontal and offset monocular views against gold-standard marker-based MOT in nine adults performing a VR-based UERW task. Key findings show frontal monocular measurements closely match the marker-based reference (mean bias ~0.6%), while offset monocular observations systematically underestimate workspace, highlighting view-dependent depth and occlusion limitations. The work demonstrates the potential for scalable, single-camera quantitative UE mobility assessments in clinical settings, while noting the need for validation in patient populations and real-time deployment.

Abstract

To validate a clinically accessible approach for quantifying the Upper Extremity Reachable Workspace (UERW) using a single (monocular) camera and Artificial Intelligence (AI)-driven Markerless Motion Capture (MMC) for biomechanical analysis. Objective assessment and validation of these techniques for specific clinically oriented tasks are crucial for their adoption in clinical motion analysis. AI-driven monocular MMC reduces the barriers to adoption in the clinic and has the potential to reduce the overhead for analysis of this common clinical assessment. Nine adult participants with no impairments performed the standardized UERW task, which entails reaching targets distributed across a virtual sphere centered on the torso, with targets displayed in a VR headset. Movements were simultaneously captured using a marker-based motion capture system and a set of eight FLIR cameras. We performed monocular video analysis on two of these video camera views to compare a frontal and offset camera configurations. The frontal camera orientation demonstrated strong agreement with the marker-based reference, exhibiting a minimal mean bias of $0.61 \pm 0.12$ \% reachspace reached per octanct (mean $\pm$ standard deviation). In contrast, the offset camera view underestimated the percent workspace reached ($-5.66 \pm 0.45$ \% reachspace reached). Conclusion: The findings support the feasibility of a frontal monocular camera configuration for UERW assessment, particularly for anterior workspace evaluation where agreement with marker-based motion capture was highest. The overall performance demonstrates clinical potential for practical, single-camera assessments. This study provides the first validation of monocular MMC system for the assessment of the UERW task. By reducing technical complexity, this approach enables broader implementation of quantitative upper extremity mobility assessment.

Monocular Markerless Motion Capture Enables Quantitative Assessment of Upper Extremity Reachable Workspace

TL;DR

This paper addresses the challenge of bringing objective upper-extremity mobility metrics into routine clinical care by validating a monocular MMC pipeline for UERW. The method combines PosePipe with MeTRAbs-ACAE for keypoint extraction, a MuJoCo-based full-body reconstruction via bilevel optimization, and a comparison of frontal and offset monocular views against gold-standard marker-based MOT in nine adults performing a VR-based UERW task. Key findings show frontal monocular measurements closely match the marker-based reference (mean bias ~0.6%), while offset monocular observations systematically underestimate workspace, highlighting view-dependent depth and occlusion limitations. The work demonstrates the potential for scalable, single-camera quantitative UE mobility assessments in clinical settings, while noting the need for validation in patient populations and real-time deployment.

Abstract

To validate a clinically accessible approach for quantifying the Upper Extremity Reachable Workspace (UERW) using a single (monocular) camera and Artificial Intelligence (AI)-driven Markerless Motion Capture (MMC) for biomechanical analysis. Objective assessment and validation of these techniques for specific clinically oriented tasks are crucial for their adoption in clinical motion analysis. AI-driven monocular MMC reduces the barriers to adoption in the clinic and has the potential to reduce the overhead for analysis of this common clinical assessment. Nine adult participants with no impairments performed the standardized UERW task, which entails reaching targets distributed across a virtual sphere centered on the torso, with targets displayed in a VR headset. Movements were simultaneously captured using a marker-based motion capture system and a set of eight FLIR cameras. We performed monocular video analysis on two of these video camera views to compare a frontal and offset camera configurations. The frontal camera orientation demonstrated strong agreement with the marker-based reference, exhibiting a minimal mean bias of \% reachspace reached per octanct (mean standard deviation). In contrast, the offset camera view underestimated the percent workspace reached ( \% reachspace reached). Conclusion: The findings support the feasibility of a frontal monocular camera configuration for UERW assessment, particularly for anterior workspace evaluation where agreement with marker-based motion capture was highest. The overall performance demonstrates clinical potential for practical, single-camera assessments. This study provides the first validation of monocular MMC system for the assessment of the UERW task. By reducing technical complexity, this approach enables broader implementation of quantitative upper extremity mobility assessment.
Paper Structure (10 sections, 7 equations, 3 figures, 3 tables)

This paper contains 10 sections, 7 equations, 3 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Methods
  3. Participants
  4. Data Collection
  5. Kinematic Processing
  6. Marker-based Processing
  7. Markerless Processing
  8. Statistical Analysis
  9. Results
  10. Discussion

Figures (3)

  • Figure 1: Methods overview. (1) Biomechanical reconstruction using two camera views: frontal and offset, compared to marker-based motion capture. (2) Video based processing included the steps descirbed in peiffer2025portable, which included post processing from PosePipeline with the MeTRAbs Bottom Up Keypoint detection, EasyMoCap Person annotation. (3) Presents the learned position optimization and biomechancial person reconstruction using MuJoCo physics engine . (4a) Marker locations placed by the same clinician for repeatability; (4b) keypoint positions from the MeTRABS detector used in analysis. (5) Workspace and octant-by-octant analysis of targets reached, with total available targets shown as red dots. Each shaded octant remains consistent throughout all figures.
  • Figure 2: Differences in percentage of reachable workspace (% Reachspace) for each octant between the marker-based system and both MMC orientations. A difference of 0 indicates perfect agreement between the MMC and the marker-based system. An * indicates a statistically significant difference in the number of targets reached between an MMC orientation and the marker-based workspace (based on pairwise comparisons with Bonferroni correction). A + indicates a statistically significant difference between the frontal and offset orientations in percentage of workspace reached.
  • Figure 3: Agreement rate between the marker-based system and both MMC orientations (frontal and offset) across all octants (Panels A & B). Agreement is expressed as a percentage, with 100% indicating perfect alignment in octant identification. Panels C and D illustrate directional error rates (Anterior–Posterior, Superior–Inferior, and Medial–Lateral) when systems were not in agreement. Error rates are shown as a percentage of total possible agreement, with 0% representing no directional error.