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 4, 2026

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
12:39

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

8.1K

Real-Time Visual Tracking with Variational Structure Attention Network.

Yeongbin Kim1, Joongchol Shin1, Hasil Park1

  • 1Department of Image, Chung-Ang University, Seoul 06974, Korea.

Sensors (Basel, Switzerland)
|November 14, 2019
PubMed
Summary

This study introduces a structure-attention network to improve visual tracking accuracy by preserving target structure during online training. The method enhances discriminative correlation filters, overcoming boundary effect issues for better performance.

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

Thermal Conductivity Reduction in Heterostructure Multilayer Composites by Phonon Density of State Mismatch.

ACS applied materials & interfaces·2026
Same author

AMST<sup>2</sup>: aggregated multi-level spatial and temporal context-based transformer for robust aerial tracking.

Scientific reports·2023
Same author

CODEN: combined optimization-based decomposition and learning-based enhancement network for Retinex-based brightness and contrast enhancement.

Optics express·2022
Same author

Photo-Realistic Image Dehazing and Verifying Networks via Complementary Adversarial Learning.

Sensors (Basel, Switzerland)·2021
Same author

Gated Dehazing Network via Least Square Adversarial Learning.

Sensors (Basel, Switzerland)·2020
Same author

SNS-CF: Siamese Network with Spatially Semantic Correlation Features for Object Tracking.

Sensors (Basel, Switzerland)·2020

Area of Science:

  • Computer Vision
  • Machine Learning
  • Artificial Intelligence

Background:

  • Discriminative correlation filters (DCF) excel in visual tracking accuracy and speed.
  • DCF methods suffer performance degradation due to target structure distortion from boundary effects in Fourier domain processing.
  • This distortion leads to learning inaccurate target representations.

Purpose of the Study:

  • To propose a novel structure-attention network to mitigate boundary effect issues in DCF-based visual tracking.
  • To enhance the robustness and adaptability of DCF trackers by preserving target structural integrity.
  • To improve the discriminative capabilities of correlation filters during online learning.

Main Methods:

  • A variational auto-encoder (VAE) is employed as a structure-attention network to generate diverse and representative target structures.
Keywords:
convolutional neural networkcorrelation filtervariational auto-encodervisual tracking

More Related Videos

Methods to Test Visual Attention Online
09:44

Methods to Test Visual Attention Online

Published on: February 19, 2015

12.4K
Gaze in Action: Head-mounted Eye Tracking of Children's Dynamic Visual Attention During Naturalistic Behavior
07:09

Gaze in Action: Head-mounted Eye Tracking of Children's Dynamic Visual Attention During Naturalistic Behavior

Published on: November 14, 2018

11.4K

Related Experiment Videos

Last Updated: Jan 4, 2026

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers
12:39

A Methodology for Capturing Joint Visual Attention Using Mobile Eye-Trackers

Published on: January 18, 2020

8.1K
Methods to Test Visual Attention Online
09:44

Methods to Test Visual Attention Online

Published on: February 19, 2015

12.4K
Gaze in Action: Head-mounted Eye Tracking of Children's Dynamic Visual Attention During Naturalistic Behavior
07:09

Gaze in Action: Head-mounted Eye Tracking of Children's Dynamic Visual Attention During Naturalistic Behavior

Published on: November 14, 2018

11.4K
  • Two denoising criteria with a novel reconstruction loss are introduced for the VAE framework to capture robust structures.
  • The proposed framework integrates these components for online training of DCF trackers.

    • Main Results:

    • The structure-attention network effectively preserves target structure, preventing distortion caused by boundary effects.
    • The proposed method demonstrates improved discriminative performance and adaptability in visual tracking.
    • Experimental results on benchmark datasets show comparable or superior performance to state-of-the-art methods.

    Conclusions:

    • The structure-attention framework significantly enhances DCF-based visual tracking by addressing boundary effect limitations.
    • The method achieves real-time processing speeds exceeding 80 frames per second.
    • This approach offers a robust solution for accurate and efficient visual tracking in challenging scenarios.