EUVIMEDEUVIMED entdecken
Crossref

Real-Time Hand Pose Estimation Using FMCW Radar on Resource-Limited Edge Devices

The Tuan Trinh, Vu Le, Tien Dat Le, Xuanque Nguyen, Minhhuy Le

IEEE Sensors Journal · 01.06.2026 · 26 (11) · 17087-17094

DOI: 10.1109/jsen.2026.3686129 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 01.06.2026. Verfasst von The Tuan Trinh und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 17 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Real-time hand pose estimation is essential for natural human–computer interaction (HCI), especially in situations where vision-based systems face challenges due to lighting, occlusion, or privacy issues. This article introduces RadarPose, an innovative framework that estimates 2-D hand keypoints for HCI applications using raw frequency-modulated continuous wave (FMCW) radar signals. To address the difficulty of manually annotating FMCW radar data, a cross-modal supervision strategy is implemented, utilizing synchronized RGB video frames processed by a pretrained vision model to generate hand keypoint labels for training the radar signals. RadarNet, the core of this framework, is a lightweight convolutional neural network (L-CNN) designed to predict 21 hand keypoints, handedness, and hand presence directly from a raw data cube. By removing the fast Fourier transform (FFT) stages, this framework facilitates end-to-end learning in the radar complex data domain, resulting in significantly reduced latency. Once trained, the proposed framework relies solely on radar data for inference, ensuring both privacy protection and invariance to lighting conditions. The model comprises only 326k parameters and achieves real-time inference in just 1.19 ms on an NVIDIA Jetson Nano. Experimental results indicate that RadarPose demonstrates competitive accuracy compared to vision-based models in well-lit environments and significantly outperforms them in low-light or occluded scenarios, highlighting its potential for robust, edge-compatible hand pose estimation. The code and dataset were available at https://github.com/thetuantrinh/UWB-Radar-Hand-Pose-Estimation.gif

Zitieren

APA

Trinh, T. T., Le, V., Le, T. D., Nguyen, X., & Le, M. (2026). Real-Time Hand Pose Estimation Using FMCW Radar on Resource-Limited Edge Devices. IEEE Sensors Journal, 26(11), 17087-17094. https://doi.org/10.1109/jsen.2026.3686129

Vancouver

Trinh TT, Le V, Le TD, Nguyen X, Le M. Real-Time Hand Pose Estimation Using FMCW Radar on Resource-Limited Edge Devices. IEEE Sensors Journal. 2026;26(11):17087-17094. doi:10.1109/jsen.2026.3686129

BibTeX
@article{Trinh2026,
  author = {Trinh, The Tuan and Le, Vu and Le, Tien Dat and Nguyen, Xuanque and Le, Minhhuy},
  title = {Real-Time Hand Pose Estimation Using FMCW Radar on Resource-Limited Edge Devices},
  journal = {IEEE Sensors Journal},
  year = {2026},
  volume = {26},
  number = {11},
  pages = {17087--17094},
  doi = {10.1109/jsen.2026.3686129}
}
Kostenlos starten →