A Novel Deep Learning Model for Human Skeleton Estimation Using FMCW Radar
Sensors · 2025
DOI: 10.3390/s25133909 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, 2025. Verfasst von Parma Hadi Rantelinggi und 3 weiteren Autor:innen. Beteiligte Einrichtung: Graduate School of Science and Technology, Keio University, Yokohama 223-8522, Japan. Hauptthemen laut MEDLINE: Deep Learning, Radar, Skeleton. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Lembaga Pengelola Dana Pendidikan, Japan Science and Technology Agency (JST) ASPIRE, Indonesia Endowment Fund for Education Agency (LPDP).
Abstract (Originalsprache)
Human skeleton estimation using Frequency-Modulated Continuous Wave (FMCW) radar is a promising approach for privacy-preserving motion analysis. However, the existing methods struggle with sparse radar point cloud data, leading to inaccuracies in joint localization. To address this challenge, we propose a novel deep learning framework integrating convolutional neural networks (CNNs), multi-head transformers, and Bi-LSTM networks to enhance spatiotemporal feature representations. Our approach introduces a frame concatenation strategy that improves data quality before processing through the neural network pipeline. Experimental evaluations on the MARS dataset demonstrate that our model outperforms conventional methods by significantly reducing estimation errors, achieving a mean absolute error (MAE) of 1.77 cm and a root mean squared error (RMSE) of 2.92 cm while maintaining computational efficiency.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Graduate School of Science and Technology, Keio University, Yokohama 223-8522, Japan.
- Department of Computer Engineering, Universitas Papua, Manokwari 98314, Indonesia.
- Faculty of Science and Technology, Keio University, Yokohama 223-8522, Japan.
ORCID-Profile: Parma Hadi Rantelinggi, Xintong Shi, Mondher Bouazizi, Tomoaki Ohtsuki
Förderung
- Lembaga Pengelola Dana Pendidikan
- Japan Science and Technology Agency (JST) ASPIRE
- Indonesia Endowment Fund for Education Agency (LPDP)
- JST ASPIRE
Zitieren
APA
Rantelinggi, P. H., Shi, X., Bouazizi, M., & Ohtsuki, T. (2025). A Novel Deep Learning Model for Human Skeleton Estimation Using FMCW Radar. Sensors. https://doi.org/10.3390/s25133909
Vancouver
Rantelinggi PH, Shi X, Bouazizi M, Ohtsuki T. A Novel Deep Learning Model for Human Skeleton Estimation Using FMCW Radar. Sensors. 2025. doi:10.3390/s25133909
BibTeX
@article{Rantelinggi2025,
author = {Rantelinggi, Parma Hadi and Shi, Xintong and Bouazizi, Mondher and Ohtsuki, Tomoaki},
title = {A Novel Deep Learning Model for Human Skeleton Estimation Using FMCW Radar},
journal = {Sensors},
year = {2025},
doi = {10.3390/s25133909}
}
EUVIMED