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

Position invariant recognition in the visual system with cluttered environments.

S M Stringer1, E T Rolls

  • 1Oxford University, Department of Experimental Psychology, UK.

Neural Networks : the Official Journal of the International Neural Network Society
|August 11, 2000
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

Brain mechanisms for invariant visual recognition and learning.

Behavioural processes·2014
Same author

Taste and olfactory status in a gourmand with a right amygdala lesion.

Neurocase·2013
Same author

Preferences for medication attributes among patients with type 2 diabetes mellitus in the UK.

Diabetes, obesity & metabolism·2013
Same author

The role of independent motion in object segmentation in the ventral visual stream: Learning to recognise the separate parts of the body.

Vision research·2011
Same author

Taste, olfactory and food texture reward processing in the brain and obesity.

International journal of obesity (2005)·2010
Same author

Continuous transformation learning of translation invariant representations.

Experimental brain research·2010

The VisNet model effectively recognizes previously learned objects in cluttered scenes but struggles with new objects against complex backgrounds. Prior exposure to relevant features aids recognition of occluded objects.

Area of Science:

  • Computational Neuroscience
  • Computer Vision
  • Artificial Intelligence

Background:

  • Invariant object recognition models, like VisNet, use temporal statistics of visual input.
  • These models associate object variations (e.g., different positions) learned in temporal proximity.
  • Cluttered environments pose challenges for object recognition systems.

Purpose of the Study:

  • Investigate the impact of cluttered environments on VisNet's object recognition performance.
  • Determine VisNet's ability to recognize previously learned vs. new objects in complex scenes.
  • Explore methods to improve VisNet's robustness in cluttered and occluded conditions.

Main Methods:

  • Utilized a hierarchical multilayer model (VisNet) with learning rules incorporating neural activity traces.

Related Experiment Videos

  • Tested VisNet's recognition of stimuli learned on plain backgrounds presented in cluttered scenes.
  • Evaluated VisNet's learning of new objects against cluttered backgrounds.
  • Extended analysis to partially occluded objects.

    • Main Results:

    • VisNet successfully recognized previously learned stimuli in cluttered environments.
    • VisNet exhibited difficulty learning new objects when presented against cluttered backgrounds.
    • Pre-existing stimulus-tuned feature-detecting neurons in early layers ameliorated learning difficulties.
    • VisNet achieved correct recognition of previously learned, partially occluded objects.

    Conclusions:

    • VisNet's performance is robust for previously learned objects in clutter.
    • Learning new objects in cluttered scenes requires mechanisms like attention or foveation.
    • Early-layer feature detectors, established through prior exposure, enhance VisNet's resilience.
    • The model demonstrates effective recognition of previously learned occluded objects.