Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Sensor Data Prediction in Missile Flight Tests.

Sensors (Basel, Switzerland)·2022

See all related articles

Search research articles

Related Experiment Video

Updated: Jan 13, 2026

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

Radar-Based Gesture Recognition Using Adaptive Top-K Selection and Multi-Stream CNNs.

Jiseop Park¹, Jaejin Jeong¹

¹Department of Electronic Engineering, Kumoh National Institute of Technology, Gumi-si 39177, Republic of Korea.

Sensors (Basel, Switzerland)

|October 29, 2025

Summary

This summary is machine-generated.

This study introduces a novel radar-based gesture recognition system. It accurately identifies hand gestures by analyzing both hand and body reflections, improving human-computer interaction.

Keywords:

FMCW radar deep learning hand gesture recognition human–computer interaction human–machine interface radar signal preprocessing

More Related Videos

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

Related Experiment Videos

Last Updated: Jan 13, 2026

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision

Published on: March 28, 2025

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

Area of Science:

Human-Computer Interaction (HCI)
Radar Signal Processing
Machine Learning

Background:

The Internet of Things (IoT) drives demand for advanced human-computer interaction (HCI).
Millimeter-wave (mmWave) frequency-modulated continuous-wave (FMCW) radar offers privacy and illumination robustness for gesture recognition, surpassing vision-based methods.
Real-world human-machine interface (HMI) scenarios involve complex reflections from hands, arms, and torsos, complicating gesture recognition.

Purpose of the Study:

To develop a robust gesture recognition system using mmWave FMCW radar.
To address the challenge of co-existing hand and body reflections in HMI environments.
To enhance the accuracy and reliability of gesture recognition in realistic settings.

Main Methods:

Proposed Adaptive Top-K Selection preprocessing method using vector entropy to preserve informative signals from hand and body reflections.
Developed a Multi-Stream EfficientNetV2 architecture integrating temporal range and Doppler trajectories.
Implemented radar-specific data augmentation and a training optimization strategy.

Main Results:

Achieved an average accuracy of 99.5% on a public FMCW gesture dataset.
Demonstrated effective capture and utilization of both hand and body reflection signals.
Validated the system's performance in realistic HMI environments with complex reflections.

Conclusions:

The proposed Adaptive Top-K Selection and Multi-Stream EfficientNetV2 architecture enable highly accurate and reliable radar-based gesture recognition.
This approach effectively handles complex reflection scenarios in real-world HMI applications.
The system offers a promising solution for advanced HCI within the growing IoT ecosystem.