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

View-based models of 3D object recognition: invariance to imaging transformations

T Vetter1, A Hurlbert, T Poggio

  • 1Max-Planck Institut für Biologische Kybernetik, Tübingen, Germany.

Cerebral Cortex (New York, N.Y. : 1991)
|May 1, 1995
PubMed
Summary

This study presents a view-based model for object recognition using regularization networks. It captures general properties of biological recognition systems, focusing on view invariance and feature synthesis.

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

Near-metameric illumination changes affect visually perceived food attributes.

Journal of the Optical Society of America. A, Optics, image science, and vision·2025
Same author

Correction: Excess of NPM-ALK oncogenic signaling promotes cellular apoptosis and drug dependency.

Oncogene·2025
Same author

An analysis of training and generalization errors in shallow and deep networks.

Neural networks : the official journal of the International Neural Network Society·2019
Same author

Missing facial parts computed by a morphable model and transferred directly to a polyamide laser-sintered prosthesis: an innovation study.

The British journal of oral & maxillofacial surgery·2011
Same author

Response from sinha and poggio.

Trends in cognitive sciences·2011
Same author

[Mucopolysaccharidoses].

Der Orthopade·2008

Area of Science:

  • Computational Neuroscience
  • Computer Vision
  • Machine Learning

Background:

  • Object recognition is a fundamental cognitive task.
  • Biological systems achieve robust object recognition despite variations in viewpoint.
  • Existing computational models often struggle with viewpoint invariance.

Purpose of the Study:

  • To describe a view-based computational model for object recognition.
  • To capture general properties of biological object recognition architectures.
  • To develop a model that achieves view invariance.

Main Methods:

  • Utilizes a regularization network (RBF-like) as the basic module.
  • Hidden units are tuned to specific object views.
  • Simulations are used to demonstrate network output and view independence.

Related Experiment Videos

Main Results:

  • The model demonstrates largely view-independent object recognition.
  • Units can represent object view components or complex features.
  • Learning involves modifying existing features and incrementally creating new ones for invariance.
  • Hierarchical structures allow for the synthesis of complex features and templates.

Conclusions:

  • The proposed view-based model offers a plausible computational framework for biological object recognition.
  • The model's architecture supports incremental learning and hierarchical feature representation.
  • Achieving view invariance is a key outcome of the model's design and learning mechanisms.