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

Object classification for human and ideal observers

Z Liu1, D C Knill, D Kersten

  • 1NEC Research Institute, Princeton, N.J.

Vision Research
|February 1, 1995
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 perception of cast shadows.

Trends in cognitive sciences·2011
Same author

Alterations in choice behavior by manipulations of world model.

Proceedings of the National Academy of Sciences of the United States of America·2010
Same author

Associative learning of scene parameters from images.

Applied optics·2010
Same author

Ideal observer perturbation analysis reveals human strategies for inferring surface orientation from texture.

Vision research·2002
Same author

Perception of 3D surface orientation from skew symmetry.

Vision research·2001
Same author

Development of infants' sensitivity to surface contour information for spatial layout.

Perception·2001
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

Researchers developed a new method using ideal observer analysis to measure human 3D object perception. This study quantifies statistical efficiency in classifying objects from noisy images, revealing insights into human visual processing.

Area of Science:

  • Cognitive Science
  • Computer Vision
  • Human Perception

Background:

  • Understanding human visual perception is crucial for developing advanced artificial intelligence systems.
  • Quantifying the efficiency of human observers in complex tasks like 3D object recognition remains a challenge.

Purpose of the Study:

  • To introduce a novel approach for measuring human ability in 3D object perception using ideal observer analysis.
  • To compute and analyze the statistical efficiency of human subjects in a 3D object classification task.

Main Methods:

  • Trained human subjects on 11 views of a randomly shaped thick wire object.
  • Assessed performance by asking subjects to identify the best match from noisy views of the learned object.
  • Calculated statistical efficiency relative to an ideal observer model.

Related Experiment Videos

Main Results:

  • Human statistical efficiency reached up to 20% relative to the actual information in stimuli.
  • Increased object regularity (e.g., symmetry) enhanced classification efficiency for novel views.
  • Higher object regularity reduced the impact of viewpoint on classification efficiency.

Conclusions:

  • The developed ideal observer analysis provides a robust measure for human 3D object perception.
  • Object regularity significantly influences classification efficiency and viewpoint dependency in human observers.
  • Human performance relative to a 2D ideal observer exceeded 100%, indicating limitations of current sub-optimal models.