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: Mar 7, 2026

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

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

10.1K

Nonlinear Deep Kernel Learning for Image Annotation.

Mingyuan Jiu, Hichem Sahbi

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |February 11, 2017
    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

    Stochastic Graphlet Embedding.

    IEEE transactions on neural networks and learning systems·2018
    Same author

    Context-dependent logo matching and recognition.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2012
    Same author

    Context-dependent kernels for object classification.

    IEEE transactions on pattern analysis and machine intelligence·2010
    Same journal

    Change-Prior-Guided Unsupervised Change Detection of Heterogeneous Remote Sensing Images.

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

    AgonicDreamer: Enhancing Multi-View Consistency in Text-to-3D Generation via Rectified Score Distillation.

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

    BiCM-Prompt: Bidirectional Cross-Modal Prompt Tuning for Class-Incremental Learning on Multisource Remote Sensing Images.

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

    GoP-based Quality Enhancement on Video Compression.

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

    Align then Tensorize: Multi-Level Consistent Anchor Graph Learning for Scalable Multi-View Clustering.

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

    Beyond Fidelity: Diverse Image Synthesis via Retrieval-Augmented Diffusion.

    IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
    See all related articles

    Deep multiple kernel learning networks offer improved performance for image annotation tasks. This novel approach uses multi-layer combinations of kernels, outperforming traditional shallow methods in complex classification challenges.

    Area of Science:

    • Machine Learning
    • Computer Vision
    • Artificial Intelligence

    Background:

    • Multiple Kernel Learning (MKL) is a standard technique for kernel design in Support Vector Classifiers.
    • Traditional MKL methods create shallow linear combinations of kernels, which are often insufficient for capturing complex similarities in high-semantic data, particularly for image annotation.
    • The limitations of shallow kernels hinder performance in challenging classification tasks.

    Purpose of the Study:

    • To redefine multiple kernels using deep multi-layer networks for enhanced feature representation.
    • To introduce a novel deep multiple kernel framework capable of learning complex data similarities.
    • To evaluate the effectiveness of deep kernel networks in image annotation tasks.

    Main Methods:

    More Related Videos

    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

    1.2K

    Related Experiment Videos

    Last Updated: Mar 7, 2026

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

    Deep Neural Networks for Image-Based Dietary Assessment

    Published on: March 13, 2021

    10.1K
    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

    1.2K
  • Proposed a recursive definition of deep multiple kernels as multi-layered combinations of nonlinear activation functions.
  • Each layer involves combinations of elementary or intermediate kernels, resulting in a positive semi-definite deep kernel.
  • Developed four frameworks for learning network weights: supervised, unsupervised, kernel-based semi-supervised, and Laplacian-based semi-supervised.

    • Main Results:

    • Deep kernel networks integrated with Support Vector Machines (SVMs) demonstrated significant performance gains over shallow kernels.
    • The proposed method showed clear advantages in image annotation tasks.
    • Experiments on ImageCLEF photo annotation, COREL5k, and Banana datasets validated the method's effectiveness.

    Conclusions:

    • Deep multiple kernel networks provide a powerful approach for learning complex similarities in high-semantic data.
    • The proposed deep kernel framework significantly enhances performance in challenging image annotation tasks.
    • The study validates the effectiveness and superiority of deep kernel networks compared to traditional shallow methods.