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

Acquisition of visual shape primitives.

Ladan Shams1, Christoph von der Malsburg

  • 1Division of Biology, California Institute of Technology, Pasadena, CA 91125, USA. ladan@caltech.edu

Vision Research
|August 10, 2002
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

The Cooperative Network Architecture: Learning Structured Networks as Representation of Sensory Patterns.

Neural computation·2026
Same author

Crossmodal interaction of flashes and beeps across time and number follows Bayesian causal inference.

Psychonomic bulletin & review·2026
Same author

Correction: Quintero et al. Changing the Tendency to Integrate the Senses. <i>Brain Sci.</i> 2022, <i>12</i>, 1384.

Brain sciences·2024
Same author

How are segmentation and binding computed and represented in the brain?

Cognitive processing·2024
Same author

BCI Toolbox: An open-source python package for the Bayesian causal inference model.

PLoS computational biology·2024
Same author

The overlooked role of unisensory precision in multisensory research.

Current biology : CB·2024
Same journal

Computational and mathematical models in vision: Quantitative approaches to understanding visual perception.

Vision research·2026
Same journal

Complex interactions between lightness, chroma, and hue in color ensemble perception.

Vision research·2026
Same journal

Driving with autism spectrum disorder: Exploring the impact of tactile hazard warnings on gaze behavior and hazard responses.

Vision research·2026
Same journal

Early visual processing in adults with ADHD: evidence from contrast sensitivity, spatial integration, and external noise.

Vision research·2026
Same journal

Pupil reflexes generate the peripheral drift illusion due to ON/OFF motion responses.

Vision research·2026
Same journal

Perceived direction of glass patterns can flip by 90°: A neural model.

Vision research·2026
See all related articles

This study introduces a novel unsupervised learning model for acquiring visual shape primitives. The algorithm learns complex object parts from experience without prior knowledge, enabling new feature discovery.

Area of Science:

  • Computational neuroscience
  • Computer vision
  • Machine learning

Background:

  • Shape primitives are key components in visual object models, supported by behavioral data.
  • Existing research focuses on detecting shape primitives, not their acquisition.
  • A gap exists in understanding how the visual system learns these fundamental representations.

Purpose of the Study:

  • To propose and validate a computational model for the self-organized, experience-based learning of shape primitives.
  • To demonstrate unsupervised learning of complex shape primitives that can serve as intermediate object representations.
  • To introduce a novel approach for simulating the acquisition of visual representations.

Main Methods:

  • Utilized synthetic gray-level objects composed of multiple parts.

Related Experiment Videos

  • Developed an algorithm where shape primitives emerge from partial matches between different objects.
  • Employed an unsupervised learning approach without any prior knowledge of pattern attributes.

    • Main Results:

    • Successfully demonstrated unsupervised learning of complex shape primitives.
    • Showcased that shape primitives emerge naturally from the recurrence of visual patterns across objects.
    • The model provides a mechanism for learning intermediate representations for diverse objects.

    Conclusions:

    • The proposed model offers the first successful unsupervised learning of complex shape primitives.
    • Experience-driven, self-organized learning is a viable mechanism for acquiring visual representations.
    • The emergence of shape primitives from partial object matches provides insights into visual system function.