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: Jan 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

1.2K

CS-Net: convolutional spider neural network for surface-EMG-based hybrid gesture recognition.

Xi Zhang1, Jiannan Chen2, Lei Liu3

  • 1Hong Kong University of Science and Technology, Hong Kong Special Administrative Region of China, People's Republic of China.

Journal of Neural Engineering
|September 26, 2025
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

Adaptive Task-Space Control for Hydraulic Excavators Based on the High-Order Fully Actuated System Approach.

IEEE transactions on cybernetics·2026
Same author

Light chain deposition disease masquerading as hypophosphatemic osteomalacia: a case report.

Frontiers in medicine·2026
Same author

Antiviral potential of black tea and blueberry extracts against monkeypox virus by targeting orthopoxvirus surface proteins.

Current research in microbial sciences·2026
Same author

Next-Generation CAR-T Therapies: From Personalized Anticancer Drugs to Broadly Applicable Disease-Modifying Treatments.

Transplantation and cellular therapy·2026
Same author

In vivo generation of CAR-T cells: Current obstacles, strategic solutions, and clinical translation.

Critical reviews in oncology/hematology·2026
Same author

Hippocampus undergoes transcriptomic changes and synaptic alterations in a Slc17a5 heterozygous mouse model with working memory deficit.

Neuroscience·2026

A novel convolutional spider neural network (CS-Net) with transfer learning (TL) achieves 90.6% accuracy for classifying hybrid gestures from surface electromyography (sEMG) signals. This method shows practical utility in real-time object grasping tasks.

Area of Science:

  • Biomedical Engineering
  • Machine Learning
  • Human-Computer Interaction

Background:

  • Surface electromyography (sEMG) signals are crucial for understanding human movement.
  • Classifying complex hybrid gestures integrating wrist postures and hand movements remains challenging.
  • Existing deep learning models require extensive labeled data for sEMG gesture recognition.

Purpose of the Study:

  • To introduce a novel Convolutional Spider Neural Network (CS-Net) architecture for hybrid gesture classification.
  • To leverage transfer learning (TL) to improve sEMG classification accuracy and generalization.
  • To evaluate the performance of CS-Net with TL on a custom hybrid gesture dataset and public databases.

Main Methods:

  • Developed a multi-stream CS-Net architecture to fuse diverse sEMG features (raw signals, FFT).
Keywords:
CNNhybrid gesture recognitiononline experimentsEMGsurface electromyographytransfer learning

More Related Videos

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

2.2K
Extraction of the EPP Component from the Surface EMG
07:16

Extraction of the EPP Component from the Surface EMG

Published on: December 16, 2009

12.9K

Related Experiment Videos

Last Updated: Jan 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

1.2K
Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping
09:41

Estimation of Contact Regions Between Hands and Objects During Human Multi-Digit Grasping

Published on: April 21, 2023

2.2K
Extraction of the EPP Component from the Surface EMG
07:16

Extraction of the EPP Component from the Surface EMG

Published on: December 16, 2009

12.9K
  • Implemented a TL strategy involving pre-training on wrist postures and fine-tuning on hybrid gestures.
  • Conducted offline experiments on a 12-hybrid gesture dataset and validated on Ninapro databases (DB1, DB4, DB5).
  • Performed real-time online experiments for object grasping tasks.

    • Main Results:

    • CS-Net with TL achieved an average accuracy of 90.6% on the custom hybrid gesture dataset.
    • Demonstrated generalization capabilities on Ninapro datasets with accuracies of 68.7% (DB1), 61.5% (DB4), and 66.3% (DB5).
    • Achieved a 90% success rate in real-time online object grasping tasks.

    Conclusions:

    • CS-Net significantly enhances sEMG classification accuracy for hybrid gestures.
    • The TL strategy further boosts performance and improves model generalization.
    • The proposed method exhibits robustness and practical utility for real-world human-computer interaction applications.