Comparison of deep learning-based three-dimensional human pose estimation methods with motion capture for gesture research
Clinical Snapshot
PICO Framework
| P — Population | Ten healthy participants performing gesture-rich speech |
| I — Intervention | Four deep learning-based human pose estimation methods (two monocular, two stereo with triangulation) |
| C — Comparator | Optical motion capture (MoCap) system as reference standard |
| O — Outcomes | Accuracy of 13 upper-body keypoints (wrists, elbows, shoulders, fingers, face) measured by Euclidean distance error and 3D gesture space overlap |
Bottom Line
This validation study demonstrates that stereo deep learning-based human pose estimation methods can achieve clinically meaningful accuracy (average error 49.4mm) compared to gold-standard motion capture for measuring 3D gestures. With 75% spatial agreement, these methods offer a cost-effective alternative to expensive motion capture systems for gesture research. The technology could democratize access to 3D gesture analysis, particularly benefiting smaller research institutions and clinical settings. However, the small sample size (n=10) and limited methodological detail in this abstract limit confidence in generalizability. While promising for research applications, clinical implementation would require validation in larger, more diverse populations and specific clinical contexts before routine adoption.
Key Findings
P Value: Statistical significance reported for stereo vs monocular comparison
Effect Size: Average error of 49.4 mm for most accurate stereo method
Primary Outcome: Accuracy of HPE methods compared to MoCap for 13 upper-body keypoints
Nnt Or Sensitivity: 75.4% spatial overlap between HPE and MoCap at 50mm voxel resolution
Confidence Interval: Not reported in abstract
Clinical Application
High feasibility - uses standard video cameras rather than expensive MoCap equipment, suitable for non-experts Could support Australian research institutions and clinical settings lacking access to expensive MoCap systems, relevant for speech pathology and communication research under NHMRC funding priorities Researchers studying human gesture and communication, potentially applicable to speech therapy and communication disorder assessment
Abstract
Humans often make gestures while speaking. These gestures have been extensively researched in psychology and cognitive science, revealing their functions and roles in communication. Despite being produced in three-dimensional (3D) space, gestures and gesture space have primarily been measured and analyzed using two-dimensional planes. Optical motion capture (MoCap) is commonly used for 3D gesture measurement. However, MoCap is expensive, requires a large space, and the attachment of retroreflective markers can interfere with the natural generation of gestures by the speaker. Recent advances in deep learning-based human pose estimation (HPE) offer promising alternatives by enabling 3D keypoint estimation from standard video cameras. This study investigates the accuracy of four HPEs-two monocular and two stereo with triangulation methods-in estimating upper-body keypoints commonly used in natural scenarios for gesture research. Ten participants were recorded performing gesture-rich speech by both a MoCap system and three video cameras. We compared 13 keypoints, including the wrists, elbows, shoulders, fingers, and face, with MoCap data, using Euclidean distance as an error metric. Statistical analyses revealed that stereo methods significantly outperformed monocular methods across all keypoints. The most accurate method achieved an average error of 49.4 mm with respect to MoCap, suggesting sufficient accuracy for practical gesture analysis. Additionally, we visualized 3D gesture spaces and found a 75.4% overlap between HPE and MoCap at a voxel size of 50 mm, indicating high spatial agreement. Our findings demonstrate that stereo HPE methods are viable, cost-effective alternatives to MoCap in gesture-related applications. Our ultimate objective is to create a toolbox that is accurate and easy to use for measuring human gestures in 3D space by exploiting the recent advances in HPE as an alternative to MoCap. Such a toolbox will be suitable for non-experts in machine learning, and these results lay the foundation for it.
References
- 1.Ienaga, N., & Sekine, K. (2026). Comparison of deep learning-based three-dimensional human pose estimation methods with motion capture for gesture research. PLoS One. https://doi.org/10.1371/journal.pone.0347288
Related Research
PloS one
Pronunciation assessment in foreign language learning: Reliability and scoring bias in human-generative AI evaluation
31 July 2026
The Behavioral and brain sciences
Large language models are not about natural language.
2 July 2026
Biomedical physics & engineering express
Intelligent integrated system for dysphagia rehabilitation and assessment
27 June 2026
This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service