EUVIMEDEUVIMED entdecken
Crossref

EdgeActNet: Edge Intelligence-Enabled Human Activity Recognition Using Radar Point Cloud

Fei Luo, Salabat Khan, Anna Li, Yandao Huang, Kaishun Wu

IEEE Transactions on Mobile Computing · 2024-05 · 23 (5) · 5479-5493

DOI: 10.1109/tmc.2023.3309938 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Mobile Computing, 2024-05. Verfasst von Fei Luo und 4 weiteren Autor:innen. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 69 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Human activity recognition (HAR) has become a research hotspot because of its wide range of application prospects. It has higher requirements for real-time and power-efficient processing. However, a large amount of data transfer between sensors and servers, and computation-intensive recognition models hinder the implementation of real-time HAR systems. Recently, edge computing has been proposed to address this challenge by moving computational and data storage resources to the sensors, rather than depending on a centralized server/cloud. In this paper, we investigated binary neural networks for edge intelligence-enabled HAR using radar point cloud. Point cloud can provide 3-dimensional spatial information, which is helpful to improve recognition accuracy. Time-series point cloud also brings challenges, such as larger data volume, 4-dimensional data processing, and more intensive computation. To tackle these challenges, we adopt the 2-dimensional histograms for point cloud multi-view processing and propose the EdgeActNet, a binary neural network for point cloud-based human activity classification on edge devices. In the evaluation, the EdgeActNet achieved the best results with average accuracies of 97.63% on the MMActivity dataset and 95.03% on the point cloud samples of the DGUHA dataset respectively; and saved $16.9\times$ 16 . 9 × memory consumption and $11.5\times$ 11 . 5 × inference time compared to its full-precision version. Our work also is the first to apply 2D histogram-based multi-view representation and BNNs for time-series point cloud classification.

Zitieren

APA

Luo, F., Khan, S., Li, A., Huang, Y., & Wu, K. (2024). EdgeActNet: Edge Intelligence-Enabled Human Activity Recognition Using Radar Point Cloud. IEEE Transactions on Mobile Computing, 23(5), 5479-5493. https://doi.org/10.1109/tmc.2023.3309938

Vancouver

Luo F, Khan S, Li A, Huang Y, Wu K. EdgeActNet: Edge Intelligence-Enabled Human Activity Recognition Using Radar Point Cloud. IEEE Transactions on Mobile Computing. 2024;23(5):5479-5493. doi:10.1109/tmc.2023.3309938

BibTeX
@article{Luo2024,
  author = {Luo, Fei and Khan, Salabat and Li, Anna and Huang, Yandao and Wu, Kaishun},
  title = {EdgeActNet: Edge Intelligence-Enabled Human Activity Recognition Using Radar Point Cloud},
  journal = {IEEE Transactions on Mobile Computing},
  year = {2024},
  volume = {23},
  number = {5},
  pages = {5479--5493},
  doi = {10.1109/tmc.2023.3309938}
}
Kostenlos starten →