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

Statistical correlations between two-dimensional images and three-dimensional structures in natural scenes.

Brian Potetz1, Tai Sing Lee

  • 1Department of Computer Science, Center for the Neural Basis of Cognition, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213, USA. bpotetz@cs.cmu.edu

Journal of the Optical Society of America. A, Optics, Image Science, and Vision
|July 19, 2003
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

Perceptual Inductive Bias Is What You Need Before Contrastive Learning.

Proceedings. IEEE Computer Society Conference on Computer Vision and Pattern Recognition·2026
Same author

Self-Attention-Based Contextual Modulation Improves Neural System Identification.

... International Conference on Learning Representations·2026
Same author

Modeling Rapid Contextual Learning in the Visual Cortex with Fast-Weight Deep Autoencoder Networks.

Proceedings of the ... AAAI Conference on Artificial Intelligence. AAAI Conference on Artificial Intelligence·2026
Same author

Computational constraints underlying shape and texture functional domain organization in macaque V4.

Cerebral cortex (New York, N.Y. : 1991)·2026
Same author

Self-Attention-Based Contextual Modulation Improves Neural System Identification.

ArXiv·2025
Same author

Large-scale calcium imaging reveals a systematic V4 map for encoding natural scenes.

Nature communications·2024

This study explores statistical correlations between range and intensity images, revealing significant relationships between image properties and surface shapes. These findings offer potential for improved computer vision and 3D reconstruction techniques.

Area of Science:

  • Computer Vision
  • Image Processing
  • Statistical Analysis

Background:

  • Statistical approaches to vision are increasingly popular.
  • Joint statistics of coregistered range and light-intensity images remain underexplored.

Purpose of the Study:

  • Investigate statistical correlations between images and their corresponding surface shapes.
  • Identify predictable properties between range and intensity image data.

Main Methods:

  • Analysis of linear properties in range images.
  • Simple computations on intensity information.
  • Exploration of cross-predictive relationships between image types.

Main Results:

  • Significant correlations (up to p = 0.45) found between linear properties of range and intensity images.

Related Experiment Videos

  • Identified which image properties are best predicted by the other.
  • Explored the structure of these correlations.

    • Conclusions:

    • Exploitable correlations exist between range and intensity image linear properties.
    • Understanding these correlations can enhance 3D reconstruction and scene understanding.
    • Further research into the structure of these correlations is warranted.