A Lightweight Radar–Camera Fusion Deep Learning Model for Human Activity Recognition
Sensors · 2026
DOI: 10.3390/s26030894 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, 2026. Verfasst von Minkyung Jeon und 1 weiteren Autor:innen. Beteiligte Einrichtung: Department of Information and Communication Engineering, Korea University of Technology and Education (KOREATECH), Cheonan-si 31253, Republic of Kore…. Hauptthemen laut MEDLINE: Radar, Deep Learning, Human Activities. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 46 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Korea Basic Science Institute, Korea Regional Innovation System & Education, Korea Regional Innovation System & Education.
Abstract (Originalsprache)
Human activity recognition in privacy-sensitive indoor environments requires sensing modalities that remain robust under illumination variation and background clutter while preserving user anonymity. To this end, this study proposes a lightweight radar–camera fusion deep learning model that integrates motion signatures from FMCW radar with coarse spatial cues from ultra-low-resolution camera frames. The radar stream is processed as a Range–Doppler–Time cube, where each frame is flattened and sequentially encoded using a Transformer-based temporal model to capture fine-grained micro-Doppler patterns. The visual stream employs a privacy-preserving 4×5-pixel camera input, from which a temporal sequence of difference frames is extracted and modeled with a dedicated camera Transformer encoder. The two modality-specific feature vectors—each representing the temporal dynamics of motion—are concatenated and passed through a lightweight fully connected classifier to predict human activity categories. A multimodal dataset of synchronized radar cubes and ultra-low-resolution camera sequences across 15 activity classes was constructed for evaluation. Experimental results show that the proposed fusion model achieves 98.74% classification accuracy, significantly outperforming single-modality baselines (single-radar and single-camera). Despite its performance, the entire model requires only 11 million floating-point operations (11 MFLOPs), making it highly efficient for deployment on embedded or edge devices.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Information and Communication Engineering, Korea University of Technology and Education (KOREATECH), Cheonan-si 31253, Republic of Korea.
ORCID-Profile: Sungmin Woo
Förderung
- Korea Basic Science Institute
- Korea Regional Innovation System & Education
- Korea Regional Innovation System & Education
Zitieren
APA
Jeon, M., & Woo, S. (2026). A Lightweight Radar–Camera Fusion Deep Learning Model for Human Activity Recognition. Sensors. https://doi.org/10.3390/s26030894
Vancouver
Jeon M, Woo S. A Lightweight Radar–Camera Fusion Deep Learning Model for Human Activity Recognition. Sensors. 2026. doi:10.3390/s26030894
BibTeX
@article{Jeon2026,
author = {Jeon, Minkyung and Woo, Sungmin},
title = {A Lightweight Radar–Camera Fusion Deep Learning Model for Human Activity Recognition},
journal = {Sensors},
year = {2026},
doi = {10.3390/s26030894}
}
EUVIMED