Biomechanically Accurate Gait Analysis: A 3d Human Reconstruction Framework for Markerless Estimation of Gait Parameters

Akila Pemasiri; Ethan Goan; Glen Lichtwark; Robert Schuster; Luke Kelly; Clinton Fookes

eess.IVMar 3, 2026

Biomechanically Accurate Gait Analysis: A 3d Human Reconstruction Framework for Markerless Estimation of Gait Parameters

Authors: Akila Pemasiri, Ethan Goan, Glen Lichtwark, Robert Schuster, Luke Kelly, Clinton Fookes

Organizations: Queensland University of Technology, Brisbane · Griffith University, Gold Coast

Abstract

This paper presents a biomechanically interpretable framework for gait analysis using 3D human reconstruction from video data. Unlike conventional keypoint based approaches, the proposed method extracts biomechanically meaningful markers analogous to motion capture systems and integrates them within OpenSim for joint kinematic estimation. To evaluate performance, both spatiotemporal and kinematic gait parameters were analysed against reference marker-based data. Results indicate strong agreement with marker-based measurements, with considerable improvements when compared with pose-estimation methods alone. The proposed framework offers a scalable, markerless, and interpretable approach for accurate gait assessment, supporting broader clinical and real world deployment of vision based biomechanics

Figures & tables

Panel of Figure 1 — Figure 1: Comparison of body landmarks obtained from the pose estimation [ 1 ] and biomechanical model [ 21 ] .

Figure 2: Overview of the gait analysis framework. The pipeline consists of five main stages: 2D pose estimation, 3D triangulation, body shape estimation and 3D reconstruction, extraction of anatomical markers, and gait parameter estimation. Gait parameters are derived both manually and through musculoskeletal modeling approaches such as OpenSim [ 4 ] .

		Stride time		Stride length		Step time		Step length
Method	2D Pose estimation method	corr	MAE	corr	MAE	corr	MAE	corr	MAE
	MMPose	0.5457	0.1061	0.2957	0.1628	0.425	0.0715	0.3234	0.1287
	YOLO	0.5439	0.0750	0.4117	0.1582	0.3142	0.0425	0.5711	0.0810
3D Pose only	OpenPose	0.5778	0.1027	0.5659	0.1243	0.5683	0.0700	0.5472	0.0475
	MMPose	0.6831	0.0754	0.4109	0.0910	0.5869	0.0312	0.3691	0.0554
	YOLO	0.6642	0.0487	0.5717	0.0811	0.5807	0.0274	0.7609	0.0430

Table 1: Correlation coefficient and Mean Absolute Error (MAE) of spatiotemporal gait parameters derived from synthetic markers and pose-estimated heel markers with those obtained from real markers.

Panel of Figure 3 — Figure 3: Comparison of kinematic parameters across gait phases.

Panel of Figure 4 — Figure 4: Averaged gait cycle waveforms of marker based method and proposed method in its best performing pose estimation method (i.e., OpenPose [ 1 ] ).

Panel of Figure 5 — Figure 5: Bland-Altman plots comparing marker based and markerless joint angle estimations from OpenPose. Red denotes the proposed method, and blue denotes the keypoint only method. The solid lines indicate the mean bias between the two methods, while the dashed lines (–) represent the 95 % limits of agreement (±1.96 SD).

Explore similar work

Markerless Motion Capture for Biomechanical Whole-Body Kinematic Estimation in Infants
May 16, 2026Divya Joshi, J. D. Peiffer, Colleen Peyton +1Markerless Motion Capture 3D Human Pose Estimation
Center for Bionic Medicine, Shirley Ryan AbilityLab, Chicago, IL · Dept. of Physical Therapy and Human Movement Science, Northwestern University, Chicago, IL · Dept. of Biomedical Engineering, Northwestern University, Evanston, IL +2
SynthGait-19K: A Physically Grounded Synthetic Video Dataset for Gait Parameter Estimation
Sep 8, 2026Soroush Mehraban, Xin Lei Lin, Vida Adeli +5Synthetic-to-Real Domain Adaptation Physical Consistency in Video Generation
1KITE Research Institute · University of Toronto · 3Vector Institute +2
Recovering Biomechanical Signals from Missing Keypoints Using Temporal Interpolation in Monocular Gait Analysis
Sep 9, 2026Shubham JariwalaTime Series Imputation Gait Analysis
Information Systems Technology and Design, Singapore University of Technology and Design, 8 Somapah Road, Singapore 487372, Singapore

May 16, 2026J/K move · Enter open · S save

Markerless Motion Capture for Biomechanical Whole-Body Kinematic Estimation in Infants

Divya Joshi, J. D. Peiffer, Colleen Peyton, R. James Cotton

Center for Bionic Medicine, Shirley Ryan AbilityLab, Chicago, IL · Dept. of Physical Therapy and Human Movement Science, Northwestern University, Chicago, IL · Dept. of Biomedical Engineering, Northwestern University, Evanston, IL+2

Markerless Motion Capture 3D Human Pose Estimation

Open paper AskPDF arXiv

arly identification of motor impairment in infancy relies on expert visual assessment of spontaneous movement, motivating the development of automated, objective alternatives. One promising approach is using computer vision, which benefits from high quality pose estimation from video. In this study, we systematically evaluated three state-of-the-art pose estimation frameworks (MeTRAbs-ACAE, SAM 3D Body, and Sapiens) on 100 videos over 13 sessions of 8 infants recorded with a multi-view markerless motion capture system. We quantified keypoint detection accuracy using reprojection error, geometric consistency, and Procrustes-aligned 3D position error, and demonstrated proof-of-concept for fitting an inverse kinematic framework to infant data. While Sapiens achieved the lowest reprojection error and highest geometric consistency of the methods evaluated (22.8 pixels and 0.82, respectively), SAM 3D Body provided the most comprehensive 3D information for kinematic reconstruction with Procrustes-aligned position errors of 19 to 28 mm. We demonstrate in a case comparison example that biomechanical models fit to SAM 3D estimates distinguish representative movement patterns in infants related to motor development, as identified by a clinical expert. Together, these findings highlight both the promise and current limitations of 3D pose estimation for infant biomechanics and establish preliminary groundwork for scalable, video-based assessment of early motor development.

Abstract

Explore similar work

Markerless Motion Capture for Biomechanical Whole-Body Kinematic Estimation in Infants

SynthGait-19K: A Physically Grounded Synthetic Video Dataset for Gait Parameter Estimation

Recovering Biomechanical Signals from Missing Keypoints Using Temporal Interpolation in Monocular Gait Analysis