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

Learning object correspondences with the observed transport shape measure.

Alain Pitiot1, Hervé Delingette, Arthur W Toga

  • 1Epidaure, INRIA, 2004 route des lucioles BP 93, 06 902 Sophia-Antipolis, France.

Information Processing in Medical Imaging : Proceedings of the ... Conference
|September 4, 2004
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

Paramagnetic rims are a promising diagnostic imaging biomarker in multiple sclerosis.

Multiple sclerosis (Houndmills, Basingstoke, England)·2022
Same author

Multi-Antigen Imaging Reveals Inflammatory DC, ADAM17 and Neprilysin as Effectors in Keloid Formation.

International journal of molecular sciences·2021
Same author

Ratlas-LH: An MRI template of the Lister hooded rat brain with stereotaxic coordinates for neurosurgical implantations.

Brain and neuroscience advances·2021
Same author

The timing mega-study: comparing a range of experiment generators, both lab-based and online.

PeerJ·2020
Same author

Single Test to ARrive at Multiple Sclerosis (STAR-MS) diagnosis: A prospective pilot study assessing the accuracy of the central vein sign in predicting multiple sclerosis in cases of diagnostic uncertainty.

Multiple sclerosis (Houndmills, Basingstoke, England)·2019
Same author

Evaluation of the Central Vein Sign as a Diagnostic Imaging Biomarker in Multiple Sclerosis.

JAMA neurology·2019

This study introduces a novel learning method for object correspondence, utilizing an observed transport measure to compute dense correspondence fields. This approach enhances the accuracy of matching diverse 1D, 2D, and 3D objects.

Area of Science:

  • Computer Vision
  • Computational Geometry
  • Machine Learning

Background:

  • Object correspondence is a fundamental problem in computer vision and geometry.
  • Existing methods often struggle with complex shapes and require significant manual input.
  • Integrating explicit knowledge into learning frameworks can improve correspondence accuracy.

Purpose of the Study:

  • To develop a novel learning-based method for solving the object correspondence problem.
  • To introduce a new local shape measure, the observed transport measure, for improved matching.
  • To demonstrate the method's effectiveness across various object dimensions and types.

Main Methods:

  • A priori learning set used to compute a dense correspondence field.
  • Introduction of the 'observed transport measure' as a local shape descriptor.

Related Experiment Videos

  • Construction of a distance matrix from the measure for problem manipulation.
  • Development of two learning strategies based on the distance matrix.

    • Main Results:

    • Successfully computed dense correspondence fields between objects.
    • Demonstrated the efficacy of the observed transport measure in matching.
    • Applied the method to match 1-D, 2-D, and 3-D objects, including anatomical structures like the corpus callosum and ventricular surfaces.

    Conclusions:

    • The proposed learning method effectively addresses the object correspondence problem by incorporating explicit knowledge.
    • The observed transport measure and distance matrix provide a robust framework for shape matching.
    • The approach shows promise for applications in computer vision, medical imaging, and geometric analysis.