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 Concept Videos

You might also read

Related Articles

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

Sort by
Same author

Src<sup>E97A</sup> Mediates Convergent Regulation of Arterial/Venous Thrombosis and Immune Homeostasis via β3-Src Dissociation.

Journal of thrombosis and haemostasis : JTH·2026
Same author

Embedded Three-Dimensional Bubble-Like Polar Textures with Ionic-Ordering-Assisted Stabilization in CuInP<sub>2</sub>S<sub>6</sub>.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

A high-resolution gridded dataset of biomass resource potentials in China with policy-aware land-use constraints for renewable energy planning.

Scientific data·2026
Same author

Predictive model of sarcopenia in chronic kidney disease: an integrated approach of bioinformatics, machine learning, and clinical validation.

Frontiers in physiology·2026
Same author

Cardiac monitoring in adolescent and young adult cancer patients treated with anthracyclines: a longitudinal descriptive study.

Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer·2026
Same author

Galectin-3 in microglia mediates neuroinflammation-induced cognitive dysfunction via selective elimination of excitatory synapses in hippocampal CA1.

Brain research·2026

Related Experiment Video

Updated: Aug 24, 2025

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

706

Improved 3D-ResNet sign language recognition algorithm with enhanced hand features.

Shiqi Wang1, Kankan Wang1, Tingping Yang1

  • 1College of Electronic and Information Engineering, Shandong University of Science and Technology, Qingdao, 266590, Shandong, China.

Scientific Reports
|October 25, 2022
PubMed
Summary
This summary is machine-generated.

This study introduces an improved 3D-ResNet algorithm for sign language recognition, enhancing hand feature detection to significantly boost accuracy. The new method achieves over 91% recognition accuracy, outperforming existing approaches.

More Related Videos

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

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

Published on: December 15, 2023

4.0K
Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility
07:46

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility

Published on: August 9, 2024

830

Related Experiment Videos

Last Updated: Aug 24, 2025

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

706
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

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

Published on: December 15, 2023

4.0K
Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility
07:46

Author Spotlight: Revolutionizing Remote Surgery with Augmented Reality and Robotics for Enhanced Precision and Accessibility

Published on: August 9, 2024

830

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Human-Computer Interaction

Background:

  • Sign language recognition faces challenges due to small hand regions, low resolution, fast motion, occlusion, and blur.
  • Existing methods often miss crucial information by relying solely on global features, limiting sign language recognition performance.

Purpose of the Study:

  • To propose an improved 3D-ResNet algorithm for sign language recognition with enhanced hand features.
  • To improve the accuracy and speed of sign language recognition by effectively capturing hand-specific details.

Main Methods:

  • Utilized an improved EfficientDet network with Bi-FPN and attention modules for robust detection of small hand regions.
  • Employed an improved 3D-ResNet18 architecture for feature extraction, incorporating a three-branch approach for global, left-hand, and right-hand image sequences.
  • Fused features from multiple branches to enhance attention to hand movements and improve representation of sign language dynamics.

Main Results:

  • The proposed algorithm achieved a Top-1 recognition accuracy of 91.12% on the CSL dataset.
  • Demonstrated superior performance compared to baseline networks (C3D, P3D, 3D-ResNet) and recent methods (I3D+BLSTM, B3D ResNet, AM-ResC3D+RCNN).
  • Evaluated using metrics like Top-N accuracy, mAP, FLOPs, and Parm, confirming the algorithm's efficiency and effectiveness.

Conclusions:

  • The enhanced hand feature detection combined with the 3D convolutional neural network significantly improves sign language recognition accuracy.
  • The proposed method effectively addresses challenges like small target detection and feature representation in sign language videos.
  • This approach offers a promising solution for more accurate and reliable sign language recognition systems.