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

Interesting objects are visually salient.

Lior Elazary1, Laurent Itti

  • 1Department of Computer Science, University of Southern California, Los Angeles, CA, USA. elazary@usc.edu

Journal of Vision
|May 20, 2008
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

Value Explicit Pretraining for Learning Transferable Representations.

IEEE robotics and automation letters·2026
Same author

Saliency Response in Superior Colliculus at the Future Saccade Goal Predicts Fixation Duration during Free Viewing of Dynamic Scenes.

The Journal of neuroscience : the official journal of the Society for Neuroscience·2024
Same author

Expert-level sleep staging using an electrocardiography-only feed-forward neural network.

Computers in biology and medicine·2024
Same author

Ferroelectric FET-based context-switching FPGA enabling dynamic reconfiguration for adaptive deep learning machines.

Science advances·2024
Same author

Eye tracking identifies biomarkers in α-synucleinopathies versus progressive supranuclear palsy.

Journal of neurology·2022
Same author

Pupillary responses to differences in luminance, color and set size.

Experimental brain research·2022
Same journal

Analysis of human visual experience data.

Journal of vision·2026
Same journal

Pyramid-based Bayesian modeling for high-resolution behavioral analysis.

Journal of vision·2026
Same journal

Sensation without perception: The white whale effect and perceptual blindness in autonomous vehicles.

Journal of vision·2026
Same journal

Gaze behavior during closed-captioned movie viewing adapts to absent audio through more frequent switching between text and scene.

Journal of vision·2026
Same journal

In pursuit of saccade awareness: Limited volitional control and minimal conscious access to catch-up saccades during smooth pursuit eye movements.

Journal of vision·2026
Same journal

Dissociable effects of element-lifetime and stimulus-duration on local and global motion processing: An equivalent noise study.

Journal of vision·2026
See all related articles

Visual saliency, a simple low-level process, significantly guides attention to interesting objects in scenes. Computational models predict salient locations that often overlap with important objects, challenging purely cognitive explanations.

Area of Science:

  • Computational Vision
  • Cognitive Neuroscience
  • Visual Attention

Background:

  • The perception of interesting objects in visual scenes is often attributed to complex cognitive processes and object recognition.
  • However, the role of simpler, low-level visual features in guiding attention remains an area of active investigation.

Purpose of the Study:

  • To investigate the contribution of low-level visual saliency to identifying interesting objects within complex visual scenes.
  • To evaluate the predictive power of a computational bottom-up attention model in locating salient regions corresponding to objects.

Main Methods:

  • Utilized the LabelMe database, comprising 24,863 photographs with 74,454 manually outlined objects.
  • Employed a computational model of bottom-up visual attention to predict salient locations within the images.

Related Experiment Videos

  • Assessed the overlap between the model's predicted salient locations and the manually outlined objects.

    • Main Results:

    • The model's top predicted salient location fell within a labeled object region in 43% of images (chance 21%).
    • One or more of the top three salient locations overlapped with an outlined object in 76% of images (chance 43%).
    • Predictive performance plateaued after the top six salient locations, indicating diminishing returns.

    Conclusions:

    • Low-level visual properties, as captured by bottom-up saliency, significantly constrain the selection of interesting objects.
    • These findings suggest that visual attention mechanisms are heavily influenced by basic visual characteristics, not solely by high-level cognitive interpretation.