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 Videos

Multiscale segmentation with vector-valued nonlinear diffusions on arbitrary graphs.

Xiaogang Dong1, Ilya Pollak

  • 1School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN 47907, USA. dongx@ecn.purdue.edu

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|July 13, 2006
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

An integer-valued spatial autoregressive model with application to COVID-19 counts.

Journal of applied statistics·2026
Same author

Litter mixture decomposition enhances the accumulation of soil active carbon and nitrogen in an alpine grassland.

Frontiers in microbiology·2026
Same author

Zooming in on methylation of homogalacturonan pectins: How degree and pattern dictate physicochemical properties, bioactivities, and potential applications.

Carbohydrate polymers·2026
Same author

The Transcription Factor TFAP2A Induces M2 Macrophage Polarization by Upregulating EPS8L3 Expression, Thereby Accelerating the Malignant Progression of Pancreatic Ductal Adenocarcinoma.

Pancreas·2026
Same author

Air quality prediction model based on deep learning hybrid framework.

Scientific reports·2026
Same author

Single-cell transcriptome sequencing reveals tumor stem cells and their molecular characteristics in intrahepatic cholangiocarcinoma.

Scientific reports·2025
Same journal

Hyperbolic Cycle Alignment for Infrared-Visible Image Fusion.

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

Learning Gaze Synthesizer via 3D-eye Controlled Diffusion and Cross-domain Feature Alignment.

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

Underlying Semantic Diffusion for Effective and Efficient In-Context Learning.

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

DiffRES: Unleashing Text-to-Image Diffusion Models for Generative Referring Expression Segmentation without Information Leakage.

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

Location Matters: Frequency-Spatial Dual Space Adaptation for Cross-Domain Few-Shot Segmentation.

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

BayeTopo: Bayesian-based Topology-guided Learning for Vascular Imaging Segmentation.

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

We introduce new nonlinear diffusion equations for segmenting multivalued images, extending prior methods. These equations effectively process complex vector-valued image data on graphs, incorporating shape information for improved segmentation results.

Area of Science:

  • Computer Vision
  • Image Processing
  • Applied Mathematics

Background:

  • Existing methods for image segmentation often struggle with multivalued and vector-valued image data.
  • Stabilized Inverse Diffusion Equations (SIDEs) have shown promise for scalar-valued image processing tasks.
  • There is a need for advanced diffusion models capable of handling complex image structures and data types.

Purpose of the Study:

  • To propose a novel family of nonlinear diffusion equations for image segmentation.
  • To extend the capabilities of diffusion models to vector-valued images and arbitrary graph structures.
  • To integrate shape information into the diffusion process for enhanced segmentation accuracy.

Main Methods:

  • Development of a new class of nonlinear diffusion equations.

Related Experiment Videos

  • Application of these equations to the segmentation of multivalued images.
  • Extension of diffusion models to process vector-valued images defined on arbitrary graphs.
  • Incorporation of novel shape-based information during the diffusion process.

    • Main Results:

    • The proposed diffusion equations are shown to be an extension of stabilized inverse diffusion equations.
    • The methods successfully process vector-valued images on arbitrary graphs, demonstrating suitability for segmentation.
    • Effective utilization of shape information during diffusion leads to improved segmentation outcomes.
    • Demonstrated effectiveness across a wide range of segmentation tasks.

    Conclusions:

    • The novel nonlinear diffusion equations provide a powerful framework for segmenting complex, multivalued images.
    • The ability to process vector-valued data on arbitrary graphs significantly broadens the applicability of diffusion-based segmentation.
    • The integration of shape information offers a promising direction for future advancements in image segmentation.