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

Dynamic binding in a neural network for shape recognition.

J E Hummel1, I Biederman

  • 1University of Minnesota, Twin Cities.

Psychological Review
|July 1, 1992
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

Mental set and mental arithmetic.

Memory & cognition·2013
Same author

Relating priming and repetition suppression.

International journal of neural systems·2002
Same author

Size tuning in the absence of spatial frequency tuning in object recognition.

Vision research·2001
Same author

Inferior temporal neurons show greater sensitivity to nonaccidental than to metric shape differences.

Journal of cognitive neuroscience·2001
Same author

The use of diagrams in analogical problem solving.

Memory & cognition·2001
Same author

Recognizing depth-rotated objects: a review of recent research and theory.

Spatial vision·2001
Same journal

Perception and action as one: Re-integrating research on human action through event files.

Psychological review·2026
Same journal

Associative learning explains "intuitive statistics" in animals.

Psychological review·2026
Same journal

A reciprocal model of practice and skill: Navigating between dropout and expertise.

Psychological review·2026
Same journal

The relative psychometric function: A general analysis framework for relating psychological processes.

Psychological review·2026
Same journal

A taxonomy of discriminatory behavior.

Psychological review·2026
Same journal

Extreme-value signal detection theory for recognition memory: The parametric road not taken.

Psychological review·2026
See all related articles

This study introduces a neural network capable of creating viewpoint-invariant structural descriptions of objects from a single image. It solves the dynamic binding problem using synchronized neural activity for efficient object recognition.

Area of Science:

  • Cognitive Science
  • Computational Neuroscience
  • Artificial Intelligence

Background:

  • Human object recognition from single views relies on viewpoint-invariant structural descriptions.
  • These descriptions specify object parts and their interrelations.
  • The dynamic binding problem hinders creating such representations.

Purpose of the Study:

  • To present a neural network model for generating viewpoint-invariant structural descriptions.
  • To demonstrate a solution to the dynamic binding problem for object representation.

Main Methods:

  • The model employs synchronized oscillatory activity among independent units.
  • Synchrony is used to parse images into constituent parts.
  • Synchrony binds attributes (parts, relations) to represent object structure.

Related Experiment Videos

Main Results:

  • The neural network successfully generates structural descriptions.
  • The model demonstrates how synchronized activity addresses dynamic binding.
  • This approach allows for economical and attribute-rich shape representations.

Conclusions:

  • Synchronized neural activity provides a viable mechanism for solving the dynamic binding problem.
  • The proposed model offers an efficient method for viewpoint-invariant object representation.
  • This work advances computational models of object recognition and structural description.