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 Video

Updated: Apr 17, 2026

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

9.7K

Applying artificial vision models to human scene understanding.

Elissa M Aminoff1, Mariya Toneva2, Abhinav Shrivastava3

  • 1Center for the Neural Basis of Cognition, Carnegie Mellon University Pittsburgh, PA, USA ; Department of Psychology, Carnegie Mellon University Pittsburgh, PA, USA.

Frontiers in Computational Neuroscience
|February 21, 2015
PubMed
Summary

Related Concept Videos

Vision01:24

Vision

61.7K
Vision is the result of light being detected and transduced into neural signals by the retina of the eye. This information is then further analyzed and interpreted by the brain. First, light enters the front of the eye and is focused by the cornea and lens onto the retina—a thin sheet of neural tissue lining the back of the eye. Because of refraction through the convex lens of the eye, images are projected onto the retina upside-down and reversed.
61.7K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

FEDERATED LEARNING OF ROBUST INDIVIDUALIZED DECISION RULES WITH APPLICATION TO HETEROGENEOUS MULTIHOSPITAL SEPSIS POPULATION.

The annals of applied statistics·2026
Same author

SNAC-DB: An ML-ready database for antibody and NANOBODY® VHH-antigen complexes with expanded structural diversity and real-world benchmarking.

Protein science : a publication of the Protein Society·2026
Same author

Realization of High-Reliable Coherent-State Quantum Secure Communication.

Research (Washington, D.C.)·2026
Same author

In vivo CAR-M therapy: advancing precision delivery and programmable immune remodeling.

Cell communication and signaling : CCS·2026
Same author

[Retracted]Oblongifolin C reverses GEM resistance via suppressing autophagy flux in bladder cancer cells.

Experimental and therapeutic medicine·2026
Same author

Evaluation of hydrological models at gauged and ungauged basins using machine learning-based limits-of-acceptability and hydrological signatures.

Journal of hydrology·2026
Same journal

Learning under constraints: a theoretical framework for comparing resource-constrained learning in biological and artificial systems.

Frontiers in computational neuroscience·2026
Same journal

MsGCN: a multi-stream graph convolutional network for multiband PLV graph fusion in EEG-based biometric identification.

Frontiers in computational neuroscience·2026
Same journal

AI-driven neuroanalytic modeling for mental health: multichannel CNN-based autism spectrum disorder detection via facial pattern analysis.

Frontiers in computational neuroscience·2026
Same journal

Modeling multiscale neural dynamics for EEG-based emotion recognition using an attentive wavelet-transformer framework.

Frontiers in computational neuroscience·2026
Same journal

New directions for complex systems in contemporary neuroscience: a morphodynamic and emergent function approach.

Frontiers in computational neuroscience·2026
Same journal

NMDA receptor kinetics drive distinct routes to chaotic firing in pyramidal neurons.

Frontiers in computational neuroscience·2026
See all related articles
This summary is machine-generated.

Artificial vision models better explain neural scene understanding than human judgments. Computer vision models incorporating mid- and high-level attributes correlate strongly with activity in scene-selective brain regions like the parahippocampal/lingual region (PPA).

Area of Science:

  • Neuroscience
  • Computer Vision
  • Cognitive Science

Background:

  • Scene understanding relies on a network of scene-selective brain regions, including the parahippocampal/lingual region (PPA), retrosplenial complex (RSC), and occipital place area (TOS).
  • Previous research often focused on single visual dimensions, neglecting the high-dimensional feature space crucial for neural representation of scenes.

Purpose of the Study:

  • To investigate how scenes are encoded in the scene-selective brain network using advanced artificial vision systems.
  • To compare the explanatory power of different computer vision models and behavioral judgments in accounting for neural activity patterns.

Main Methods:

  • Correlated similarity matrices derived from BOLD activity in scene-selective regions with those from behavioral judgments and various computer vision models.
Keywords:
computer visionparahippocampal place arearetrosplenial cortexscene processingtransverse occipital sulcus

Related Experiment Videos

Last Updated: Apr 17, 2026

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

9.7K
  • Evaluated models based on their ability to capture neural patterns, particularly focusing on mid- and high-level scene attributes.

    • Main Results:

    • Computer vision models utilizing mid- and high-level scene attributes demonstrated the highest correlations with neural activity in the scene-selective network.
    • The NEIL and SUN models best explained activity in the PPA and TOS, while the GIST model was optimal for the RSC.
    • The top-performing models surpassed behavioral judgments in explaining neural data variance.
    • The NEIL model showed significant correlations across all three regions and was a top performer for PPA and TOS.

    Conclusions:

    • Artificial vision systems, especially those incorporating learned statistical regularities from large datasets (like NEIL), offer a powerful tool for understanding neural scene encoding.
    • These findings represent a significant advancement in developing detailed models of neural scene understanding and clarifying the functional roles within the scene-selective brain network.