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

Shape registration in implicit spaces using information theory and free form deformations.

Xiaolei Huang1, Nikos Paragios, Dimitris N Metaxas

  • 1Center for Computational Biomedicine Imaging and Modeling, Division of Computer and Information Sciences, Rutgers University, New Brunswick, NJ 08854-8019, USA. xiaolei@cs.rutgers.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|August 5, 2006
PubMed
Summary

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

Nuclear condensates formed by truncated mutant NEK1s impede ribosomal RNA biogenesis and drive motor dysfunction.

Nature communications·2026
Same author

Structural insights into the MLH1-FAN1 interaction reveal an uncharacterized binding interface on MLH1.

Nature communications·2026
Same author

DeepTriage-CN: integrating clinical text with vital signs for emergency department admission prediction in an aging population.

Scientific reports·2026
Same author

Interfacial Charge Transfer Activates Graphene Shells for High-Efficiency and Durable Vanadium-Cerium Redox Flow Batteries.

Small (Weinheim an der Bergstrasse, Germany)·2026
Same author

Tephritid26: A standardized, multi-angle image dataset of quarantine-significant true fruit flies for deep learning-based identification.

Scientific data·2026
Same author

Metabolic determinants of cancer immunotherapy outcomes identified by plasma profiling.

Nature medicine·2026

This study introduces a novel statistical shape registration method. It accurately aligns complex shapes, preserving topology and handling noise, occlusion, and missing parts effectively.

Area of Science:

  • Computer Vision
  • Medical Image Analysis
  • Computational Geometry

Background:

  • Shape registration is crucial for comparing and analyzing geometric data.
  • Existing methods often struggle with complex topologies, arbitrary dimensions, or non-rigid deformations.
  • Robustness to noise, occlusion, and missing data remains a significant challenge.

Purpose of the Study:

  • To develop a novel, variational, and statistical approach for robust shape registration.
  • To handle shapes of arbitrary dimensions and topologies, including those with multiple parts or open/closed structures.
  • To achieve accurate global and local non-rigid registration while preserving shape topology.

Main Methods:

  • Implicit embedding of shapes in a higher-dimensional distance transform space.

Related Experiment Videos

  • Hierarchical registration: Mutual Information for global alignment, followed by B-spline Incremental Free Form Deformations (IFFD) for local non-rigid refinement.
  • Minimization of Sum-of-Squared-Differences (SSD) for dense local registration field recovery.

    • Main Results:

    • The framework successfully registers shapes of arbitrary dimensions (2D, 3D+) and topologies.
    • Demonstrated robustness to noise, severe occlusion, and missing parts through empirical validation.
    • Generated smooth, continuous, one-to-one correspondences for local registration fields, preserving topology.

    Conclusions:

    • The proposed hierarchical shape registration framework offers a robust and versatile solution.
    • It excels in handling complex geometric shapes and challenging real-world scenarios.
    • Applications in statistical modeling of anatomical structures and 3D face analysis demonstrate its practical utility.