EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Method for Human Pose Estimation and Joint Angle Computation Through Deep Learning

Ludovica Ciardiello, Patrizia Agnello iD, Marta Petyx, Fabio Martinelli, Mario Cesarelli, Antonella Santone, Francesco Mercaldo iD

Journal of Imaging · 06.04.2026 · 12 (4) · 157

DOI: 10.3390/jimaging12040157 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Imaging, online erschienen am 06.04.2026. Verfasst von Ludovica Ciardiello und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Medicine and Health Sciences "Vincenzo Tiberio", University of Molise, 86100 Campobasso, Italy. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NextGenerationEU projects, e-DAI (Digital ecosystem for integrated analysis of heterogeneous health data related to high-impact diseases: innovative model of c….

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Human pose estimation is a crucial task in computer vision with widespread applications in healthcare, rehabilitation, sports, and remote monitoring. In this paper, we propose a deep learning-based method for automatic human pose estimation and joint angle computation, tailored specifically for physiotherapy and telemedicine scenarios. Beyond pose estimation, the proposed method is able to compute angles between joints, enabling analysis of body alignment and posture. The proposed approach is built upon a customized skeleton with 25 anatomical keypoints and a dataset composed of over 150,000 annotated and augmented images derived from multiple open-source datasets. Experimental results demonstrate the effectiveness of the proposed method, achieving a mAP@50 of 0.58 for keypoint localization and 0.98 for object detection. Moreover, we demonstrate several real-world practical use cases in evaluating exercise correctness and identifying postural deviations by exploiting the proposed method, confirming that the proposed method can represent a promising approach for automated motion analysis, with potential impact on digital health, rehabilitation support, and remote patient care.

Schlagwörter

Stichwörter der Autor:innen

HPEangle computationartificial intelligencedeep learninghuman pose estimationobject detection

Autor:innen und Institutionen

  • Department of Medicine and Health Sciences "Vincenzo Tiberio", University of Molise, 86100 Campobasso, Italy.
  • Istituto Nazionale per l'Assicurazione Contro gli Infortuni sul Lavoro, 00144 Roma, Italy.
  • Institute for High Performance Computing and Networking, National Research Council of Italy (CNR), 87036 Rende, Italy.
  • Department of Engineering, University of Sannio, 82100 Benevento, Italy.

ORCID-Profile: Patrizia Agnello, Francesco Mercaldo

Förderung

  • NextGenerationEU projects, e-DAI (Digital ecosystem for integrated analysis of heterogeneous health data related to high-impact diseases: innovative model of c…

Zitieren

APA

Ciardiello, L., Agnello, P., Petyx, M., Martinelli, F., Cesarelli, M., Santone, A., & Mercaldo, F. (2026). A Method for Human Pose Estimation and Joint Angle Computation Through Deep Learning. Journal of Imaging, 12(4), 157. https://doi.org/10.3390/jimaging12040157

Vancouver

Ciardiello L, Agnello P, Petyx M, Martinelli F, Cesarelli M, Santone A, et al. A Method for Human Pose Estimation and Joint Angle Computation Through Deep Learning. Journal of Imaging. 2026;12(4):157. doi:10.3390/jimaging12040157

BibTeX
@article{Ciardiello2026,
  author = {Ciardiello, Ludovica and Agnello, Patrizia and Petyx, Marta and Martinelli, Fabio and Cesarelli, Mario and Santone, Antonella and Mercaldo, Francesco},
  title = {A Method for Human Pose Estimation and Joint Angle Computation Through Deep Learning},
  journal = {Journal of Imaging},
  year = {2026},
  volume = {12},
  number = {4},
  pages = {157},
  doi = {10.3390/jimaging12040157}
}
Kostenlos starten →