Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 16, 2026

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

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

Published on: March 28, 2025

Efficient material selection for training occluded mmWave radar-based gesture recognisers.

Nuwan T Attygalle1, Luis A Leiva2, Matjaž Kljun3,4

  • 1Université catholique de Louvain, Louvain, Belgium. nuwan.attygalle@uclouvain.be.

Scientific Reports
|June 13, 2026
PubMed
Summary

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

Cognitive state monitoring for neuroadaptive information visualization.

Frontiers in human neuroscience·2026
Same author

Evaluating the sensitivity of dry and gel-based wearable EEG for cognitive load estimation.

Biomedical physics & engineering express·2026
Same author

Predicting fixations and gaze location from EEG.

Medical & biological engineering & computing·2025
Same author

Brain Signatures of Time Perception in Virtual Reality.

IEEE transactions on visualization and computer graphics·2025
Same author

Brainsourcing for temporal visual attention estimation.

Biomedical engineering letters·2025
Same author

Author Correction: On-device query intent prediction with lightweight LLMs to support ubiquitous conversations.

Scientific reports·2024
Same journal

Correction: A method for supervoxel-wise association studies of age and other non-imaging variables from coronary computed tomography angiograms.

Scientific reports·2026
Same journal

Poly(bromophenol blue)/CoSn(OH)<sub>6</sub> cubic particles modified pencil graphite electrode for electrochemical determination of diphenhydramine.

Scientific reports·2026
Same journal

Dietary Chlorella, Spirulina, and acidifier modulate jejunal cytokine-related gene expression in broiler chickens.

Scientific reports·2026
Same journal

Perceived physical activity barriers in university students: associations with fatigue and eating behaviours.

Scientific reports·2026
Same journal

Refuge limitation structures habitat use in agricultural landscapes: evidence from Sunda pangolins.

Scientific reports·2026
Same journal

Lightweight stateless transaction verification with outsourced witness updates for UTXO blockchains.

Scientific reports·2026
See all related articles
This summary is machine-generated.

Efficient radar gesture recognition is possible with a small, representative set of materials. Quota sampling with 14 materials achieved high accuracy, outperforming data augmentation and reducing collection needs.

Area of Science:

  • Human-Computer Interaction
  • Machine Learning
  • Sensor Technology

Background:

  • Radar-based gesture recognition offers unobtrusive interaction, capable of sensing through opaque materials.
  • Integrating radar into everyday objects requires efficient training models for diverse material conditions.
  • Current methods lack clarity on optimizing model training for robust performance across various materials.

Purpose of the Study:

  • To investigate efficient strategies for training radar-based gesture recognition models through diverse materials.
  • To compare the effectiveness of material-sampling and data-augmentation techniques.
  • To determine the optimal balance between performance and practicality in model training.

Main Methods:

  • Collected a dataset of 17,520 gesture recordings across 73 everyday materials.
Keywords:
Gesture recognitionMaterial selectionMillimetre-wave radarSamplingSynthetic noise

Related Experiment Videos

Last Updated: Jun 16, 2026

Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
08:15

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

Published on: March 28, 2025

  • Evaluated material-sampling strategies (including Quota sampling) and data-augmentation.
  • Compared classifier performance trained on full datasets, sampled subsets, and augmented data.

    • Main Results:

    • Models trained on 14 quota-sampled materials achieved high accuracies (95.8% and 91.2%), comparable to training on all 73 materials (96.8% and 91.6%).
    • Quota sampling offered the best performance-practicality trade-off.
    • Classifiers trained on augmented data performed worse than those trained on actual material-specific data.

    Conclusions:

    • A small, representative subset of real materials is sufficient for training robust radar gesture recognition models.
    • Quota sampling is a practical and effective strategy for reducing data collection efforts.
    • Carefully selected real-material data is superior to augmented data for preserving generalization in radar gesture recognition.