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: May 14, 2026

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

Learning dynamic hybrid Markov random field for image labeling.

Quan Zhou1, Jun Zhu, Wenyu Liu

  • 1Department of Electronics and Information Engineering, Huazhong University of Science and Technology, Wuhan 430074, China. qzhou.lhi@gmail.com

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|February 16, 2013
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

A non-parametric method for building predictive genetic tests on high-dimensional data.

Human heredity·2011
Same author

Predictive genes in adjacent normal tissue are preferentially altered by sCNV during tumorigenesis in liver cancer and may rate limiting.

PloS one·2011
Same author

Glycerol-3-phosphate acyltransferase 6 (GPAT6) is important for tapetum development in Arabidopsis and plays multiple roles in plant fertility.

Molecular plant·2011
Same author

Telmisartan, ramipril, or both in high-risk Chinese patients: analysis of ONTARGET China data.

Chinese medical journal·2011
Same author

Geographical detector-based risk assessment of the under-five mortality in the 2008 Wenchuan earthquake, China.

PloS one·2011
Same author

CardioDetect rapid test for the diagnosis of early acute myocardial infarction.

Journal of immunoassay & immunochemistry·2011

This study introduces a dynamic hybrid Markov random field (DHMRF) for image labeling, integrating shape, color, and texture for improved accuracy. The DHMRF model offers efficient and reliable image labeling by combining visual appearance and object shape information.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Artificial Intelligence

Background:

  • Image labeling tasks increasingly require sophisticated methods to integrate diverse visual cues.
  • Traditional approaches often rely on implicit shape information, limiting labeling accuracy.
  • There is a need for models that explicitly combine object shape with low-level visual features like color and texture.

Purpose of the Study:

  • To present a novel Dynamic Hybrid Markov Random Field (DHMRF) model for enhanced image labeling.
  • To explicitly integrate middle-level object shape information with low-level visual appearance features.
  • To develop an efficient and accurate image labeling method that outperforms existing techniques.

Main Methods:

  • Developed a DHMRF model where nodes represent deformable templates or appearance models (visual prototypes).

Related Experiment Videos

Last Updated: May 14, 2026

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

  • Edges in the DHMRF encode co-occurrence and spatial layered context between labels and prototypes.
  • Designed an iterative algorithm with a branch-and-bound schema for efficient feature selection and parameter estimation.

    • Main Results:

    • The DHMRF model seamlessly integrates color, texture, and shape cues for more accurate and reliable image labeling.
    • Achieved high computational efficiency in model learning and parameter estimation.
    • Demonstrated superior performance over state-of-the-art methods on the MSRC 21-class and lotus hill institute 15-class datasets in terms of recognition accuracy and implementation efficiency.

    Conclusions:

    • The proposed DHMRF model effectively combines explicit shape information with visual appearance for superior image labeling.
    • The iterative learning algorithm ensures computational efficiency and accurate parameter estimation.
    • DHMRF represents a significant advancement in image labeling, offering improved accuracy and efficiency.