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

Representation and detection of deformable shapes.

Pedro F Felzenszwalb1

  • 1Department of Computer Science, The University of Chicago, Chicago, IL 60637, USA. pff@cs.uchicago.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|February 4, 2005
PubMed
Summary

This study introduces a novel representation for deformable shapes, enabling efficient and robust detection in images. The new method overcomes challenges in nonrigid matching, offering optimal solutions even in cluttered scenes.

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

Automated brightfield layerwise evaluation in three-dimensional micropatterning via two-photon polymerization.

Optics express·2024
Same author

Fast Inference with Min-Sum Matrix Product.

IEEE transactions on pattern analysis and machine intelligence·2011
Same author

Dynamic programming and graph algorithms in computer vision.

IEEE transactions on pattern analysis and machine intelligence·2010
Same author

Object detection with discriminatively trained part-based models.

IEEE transactions on pattern analysis and machine intelligence·2010

Area of Science:

  • Computer Vision
  • Image Analysis
  • Computational Geometry

Background:

  • Detecting deformable shapes in images is challenging due to the vast number of possible nonrigid transformations.
  • Existing deformable template models struggle with finding optimal matches efficiently.
  • High computational complexity and sensitivity to initialization limit current methods.

Purpose of the Study:

  • To develop an efficient and robust method for representing and detecting deformable shapes in images.
  • To address the limitations of traditional deformable template models in nonrigid matching.
  • To introduce a new shape representation that facilitates global optimal solutions.

Main Methods:

  • A novel shape representation based on triangulated polygons is proposed.

Related Experiment Videos

  • An efficient matching algorithm is developed to find global optimal solutions for nonrigid transformations.
  • The algorithm minimizes a broad class of energy functions.
  • A method for learning nonrigid shape models from examples is presented.

    • Main Results:

    • The proposed method achieves efficient global optimal solutions for nonrigid matching.
    • Experimental results demonstrate robust shape detection in medical images and natural scenes, even with high clutter.
    • The method is independent of initialization and yields accurate matches.
    • Effective learning of constrained nonrigid shape models from example data was achieved.

    Conclusions:

    • The new triangulated polygon representation and matching algorithm significantly advance deformable shape detection.
    • This approach offers a robust and efficient solution for nonrigid image analysis tasks.
    • The ability to learn shape models enhances the applicability of the method to various object classes.