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 viewpoint invariant perceptual representations from cluttered images.

Michael W Spratling1

  • 1Division of Engineering, King's College, London, UK. michael.spratling@kcl.ac.uk

IEEE Transactions on Pattern Analysis and Machine Intelligence
|May 7, 2005
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

Cellular psychology: relating cognition to context-sensitive pyramidal cells.

Trends in cognitive sciences·2024
Same author

Comprehensive assessment methods are key to progress in deep learning.

The Behavioral and brain sciences·2023
Same author

Shape-Texture Debiased Training for Robust Template Matching.

Sensors (Basel, Switzerland)·2022
Same author

Distinguishing theory from implementation in predictive coding accounts of brain function.

The Behavioral and brain sciences·2013
Same author

Image segmentation using a sparse coding model of cortical area V1.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2012
Same author

A model of partial reference frame transforms through pooling of gain-modulated responses.

Cerebral cortex (New York, N.Y. : 1991)·2012

This study enhances object recognition by improving how perceptual representations are learned. A modified learning method allows for more robust generalization of object features, even when multiple objects appear together.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Cognitive Science

Background:

  • Object recognition requires specific yet flexible perceptual representations.
  • Current methods for viewpoint-invariant learning often fail with co-occurring objects.
  • Real-world visual input frequently contains multiple objects simultaneously.

Purpose of the Study:

  • To propose a modification to existing learning methods for perceptual representations.
  • To enable robust learning of invariant representations in the presence of multiple objects.
  • To improve the generalization capabilities of object recognition systems.

Main Methods:

  • A standard method involving temporal associations across image sequences was adapted.
  • A simple modification was introduced to the learning process.

Related Experiment Videos

  • The modified method was evaluated for its ability to handle co-occurring stimuli.

    • Main Results:

    • The proposed modification overcomes the limitation of requiring isolated stimuli.
    • More robust learning of invariant representations was achieved.
    • The method shows improved performance in complex visual scenes.

    Conclusions:

    • The modified learning approach enhances the robustness of perceptual representations.
    • This advancement is crucial for real-world object recognition applications.
    • The findings contribute to more effective machine vision systems.