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

CONFIGR: a vision-based model for long-range figure completion.

Gail A Carpenter1, Chaitanya Sai Gaddam, Ennio Mingolla

  • 1Department of Cognitive and Neural Systems, Boston University, 677 Beacon Street, Boston, MA 02215, USA.

Neural Networks : the Official Journal of the International Neural Network Society
|November 21, 2007
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

Psychophysics of neon color spreading: Chromatic and temporal factors are not limiting.

Vision research·2024
Same author

Extreme image transformations affect humans and machines differently.

Biological cybernetics·2023
Same author

Looking to the future: Learning from experience, averting catastrophe.

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

Looking to the future: Learning from experience, averting catastrophe.

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

Mitigation of Effects of Occlusion on Object Recognition with Deep Neural Networks through Low-Level Image Completion.

Computational intelligence and neuroscience·2016
Same author

Tuning Properties of MT and MSTd and Divisive Interactions for Eye-Movement Compensation.

PloS one·2015
Same journal

Anchor-based disentanglement framework for incremental multi-view clustering.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Complex-valued amplitude-phase interference modeling for adversarially robust classification.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

TraNce: Type-aware hypergraph neural network with biological mediators for drug repositioning.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Decentralized ADMM for factorization-based Low-rank matrix estimation.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Memristive neuromorphic circuit design inspired by the neural mechanisms of conditioned fear.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Q-learning based asynchronous Boolean control networks stabilization with data loss.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

CONFIGR (CONtour FIgure GRound) is a computational model that reconstructs incomplete images by filling in missing contours and connecting sparse elements. This biologically inspired vision model enhances object recognition by completing figures and distinguishing them from the ground.

Area of Science:

  • Computational Vision
  • Artificial Intelligence
  • Cognitive Science

Background:

  • Biological vision systems excel at completing incomplete figures and segmenting objects.
  • Existing computational models often struggle with long-range image completion and distinguishing figure from ground.

Purpose of the Study:

  • To introduce CONFIGR (CONtour FIgure GRound), a novel computational model for image completion based on biological vision principles.
  • To demonstrate CONFIGR's ability to reconstruct sparse and noisy images for improved object recognition.

Main Methods:

  • CONFIGR employs a two-stage process: initial figure identification followed by long-range completion via filling-in.
  • The model operates on a single, fixed spatial scale defined by pixel size, with transparent design principles.

Related Experiment Videos

  • It balances figure completion with ground segregation to prevent spurious reconstructions.

    • Main Results:

    • CONFIGR successfully completes missing contours, connects sparse dots, and unifies occluded objects.
    • The model self-scales completion distances, handling gaps of varying lengths and dense figure groups.
    • Simulations illustrate the model's multi-scale capabilities and analytical derivation.

    Conclusions:

    • CONFIGR provides a robust framework for image reconstruction, inspired by biological vision.
    • The model's ability to perform long-range image completion is crucial for adaptive processors and image recognition tasks.
    • Open-source availability facilitates further research and application of CONFIGR.