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: Nov 11, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

760

Improved deep CNNs based on Nonlinear Hybrid Attention Module for image classification.

Nan Guo1, Ke Gu1, Junfei Qiao1

  • 1Beijing Key Laboratory of Computational Intelligence and Intelligent System, Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China.

Neural Networks : the Official Journal of the International Neural Network Society
|March 25, 2021
PubMed
Summary
This summary is machine-generated.

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

Characterization of Nasopharyngeal Microbiota Dysbiosis in Children with <i>Mycoplasma pneumoniae</i> Pneumonia.

Microorganisms·2026
Same author

Perceptual Quality Assessment of Low-Light Enhanced Images: A Multi-Annotated Subjective Dataset and a Multimodal Objective Method.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

Clinical features and efficacy analysis of idiopathic sudden sensorineural hearing loss in children: a single-center, retrospective study (2015-2025).

European archives of oto-rhino-laryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS) : affiliated with the German Society for Oto-Rhino-Laryngology - Head and Neck Surgery·2026
Same author

High temperature Néel skyrmions in simple ferromagnets.

Nature communications·2026
Same author

Hierarchical organization of tau topography across the Alzheimer's disease continuum integrates amyloid, connectome, and intrinsic molecular vulnerability.

Communications biology·2026
Same author

Histopathological differences between antrochoanal polyps and chronic rhinosinusitis with nasal polyps in pediatric patients.

World journal of pediatric surgery·2026

Researchers developed a new attention module for convolutional neural networks, enhancing training efficiency and performance. This lightweight module integrates channel and spatial attention, offering flexibility and adaptability for various deep learning architectures.

Area of Science:

  • Computer Science
  • Artificial Intelligence
  • Machine Learning

Background:

  • Attention mechanisms have significantly improved feed-forward convolutional neural networks.
  • Existing attention modules like SE and CBAM offer valuable functionalities but can be further optimized.

Purpose of the Study:

  • To design a novel, lightweight, and general-purpose attention module for convolutional neural networks.
  • To enhance network fine-tuning by combining channel and spatial attention features.
  • To create a flexible and adaptable attention module with adjustable parameters.

Main Methods:

  • A nonlinear hybrid method was proposed to combine channel and spatial attention feature maps.
  • The proposed attention module was integrated into existing deep architectures.
Keywords:
Convolutional neural networksFeature map combinationGeneral moduleHybrid attention mechanism

More Related Videos

Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.6K

Related Experiment Videos

Last Updated: Nov 11, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

760
Deep Neural Networks for Image-Based Dietary Assessment
13:19

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

9.6K
  • Experiments were conducted on CIFAR10, CIFAR100, and Fashion MNIST datasets.

    • Main Results:

    • The integrated attention module improved training efficiency and network performance compared to state-of-the-art methods.
    • The module demonstrated effectiveness in enhancing existing deep architectures with minimal computational overhead.
    • The proposed module was shown to be a more general case compared to SE and CBAM modules.

    Conclusions:

    • The novel attention module offers a flexible and efficient way to boost the performance of convolutional neural networks.
    • Its adaptability and ease of integration make it a valuable tool for improving deep learning models.
    • The module represents an advancement in attention mechanism design for computer vision tasks.