EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

SparseTrack: A Physics-Informed Transformer Framework for Real-Time Human Motion Reconstruction from Sparse IMUs

Adithya Balasubramanyam iD, Suchir Murali Velpanur iD, Sushma Edhala Jeevarathnam iD, Tejasree Chekuri Jayachandra iD, Prasad Honnavalli iD, Gowri Srinivasa iD

Sensors · 21.05.2026 · 26 (10) · 3262

DOI: 10.3390/s26103262 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 21.05.2026. Verfasst von Adithya Balasubramanyam und 5 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Engineering, PES University, Bengaluru 560085, India. Hauptthemen laut MEDLINE: Motion Capture, Movement. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Wearable inertial measurement units are widely used for human motion analysis due to their portability; however, most IMU-based motion capture systems rely on dense sensor configurations that increase cost, complexity, and usability challenges in real-world applications. To address this limitation, this paper presents a sparse inertial human motion reconstruction framework that uses only five wearable sensors while maintaining real-time performance and biomechanical plausibility. The proposed framework integrates Movella Xsens DOT IMUs with a learning-based inverse kinematics pipeline and a real-time biomechanical digital twin for motion reconstruction and visualization. The evaluation was conducted in two phases: first, a real-time motion streaming system was established to validate sensor alignment, coordinate frame consistency, and end-to-end latency; second, a sparse inference framework was trained using the Virginia Tech Natural Motion Dataset combined with a custom dataset containing hard negative samples. Experimental results show that the system can accurately reconstruct full-body human motion, excluding head movement, with a local Mean Per-Joint Position Error of 5.96 cm using only five sensors. Comparative ablation studies demonstrate that Transformer-based temporal modeling achieves better geometric accuracy and temporal smoothness than recurrent and convolutional baselines, while physics-informed regularization and hard negative mining significantly improve biomechanical consistency and reduce motion jitter. Real-time experiments further demonstrate that the framework operates within interactive latency limits, highlighting its potential for biomechanical digital twin applications.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Motion CaptureMovementHumansAlgorithmsBiomechanical PhenomenaWearable Electronic Devices

Stichwörter der Autor:innen

biomechanical constraintsdigital twinshuman motion reconstructionreal-time motion capturesparse inertial sensingtransformers

Autor:innen und Institutionen

  • Department of Computer Science and Engineering, PES University, Bengaluru 560085, India.

ORCID-Profile: Adithya Balasubramanyam, Suchir Murali Velpanur, Sushma Edhala Jeevarathnam, Tejasree Chekuri Jayachandra, Prasad Honnavalli, Gowri Srinivasa

Zitieren

APA

Balasubramanyam, A., Velpanur, S. M., Jeevarathnam, S. E., Jayachandra, T. C., Honnavalli, P., & Srinivasa, G. (2026). SparseTrack: A Physics-Informed Transformer Framework for Real-Time Human Motion Reconstruction from Sparse IMUs. Sensors, 26(10), 3262. https://doi.org/10.3390/s26103262

Vancouver

Balasubramanyam A, Velpanur SM, Jeevarathnam SE, Jayachandra TC, Honnavalli P, Srinivasa G. SparseTrack: A Physics-Informed Transformer Framework for Real-Time Human Motion Reconstruction from Sparse IMUs. Sensors. 2026;26(10):3262. doi:10.3390/s26103262

BibTeX
@article{Balasubramanyam2026,
  author = {Balasubramanyam, Adithya and Velpanur, Suchir Murali and Jeevarathnam, Sushma Edhala and Jayachandra, Tejasree Chekuri and Honnavalli, Prasad and Srinivasa, Gowri},
  title = {SparseTrack: A Physics-Informed Transformer Framework for Real-Time Human Motion Reconstruction from Sparse IMUs},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {10},
  pages = {3262},
  doi = {10.3390/s26103262}
}
Kostenlos starten →