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

Facial recognition using simulated prosthetic pixelized vision.

Robert W Thompson1, G David Barnett, Mark S Humayun

  • 1Lions Vision Research and Rehabilitation Center, Wilmer Ophthalmological Institute, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205, USA.

Investigative Ophthalmology & Visual Science
|October 28, 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

Low Vision Assessment in Virtual Reality (LVAVR): Spatial Computing in Virtual Reality and Real-World.

2026 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops·2026
Same author

Gamification Enhances User Engagement and Task Performance in Prosthetic Vision Testing.

Translational vision science & technology·2026
Same author

Administration of Topical NorLeu<sup>3</sup>Angiotensin(1-7) Minimizes Fibrotic Corneal Healing in Stellate Wound: A 28-Day Study.

International journal of molecular sciences·2026
Same author

Autonomous multisensory enhancement of a visual neuroprosthesis for navigation: technical proof-of-concept with simulated prosthetic vision and single-subject case study of a visual prosthesis user.

Journal of neural engineering·2026
Same author

Gamification Enhances User Engagement and Task Performance in Prosthetic Vision Testing.

medRxiv : the preprint server for health sciences·2026
Same author

Value of Treatment for Combat Ocular Trauma.

Military medicine·2025
Same journal

Impact of Subretinal Drusenoid Deposits on Ellipsoid Zone-Related Thickness Metrics.

Investigative ophthalmology & visual science·2026
Same journal

Proteomic Profiling of Optic Nerves From SMOX-Deficient Mice Identifies Regulators of Neuroinflammation and Axonal Damage in Optic Neuritis.

Investigative ophthalmology & visual science·2026
Same journal

Aflibercept and Faricimab Equipotently Restore Endothelial Barrier Function.

Investigative ophthalmology & visual science·2026
Same journal

Spatial Decomposition of Longitudinal RNFL Maps Reveals Distinct Modes of Glaucomatous Progression With Structure-Function and Genetic Signatures.

Investigative ophthalmology & visual science·2026
Same journal

The CXXC1-IGFBP6 Axis Maintains Corneal Epithelial Differentiation via H3K4me3-Dependent Transcriptional Activation.

Investigative ophthalmology & visual science·2026
Same journal

Archetypal Visual Field Analysis of Patients With Chronic Leber Hereditary Optic Neuropathy in Relation to Visual Recovery.

Investigative ophthalmology & visual science·2026
See all related articles

Facial recognition is possible with simulated pixelized vision, even with crude visual prostheses. Learning and practice significantly improve performance with different phosphene and grid parameters.

Area of Science:

  • Ophthalmology
  • Biomedical Engineering
  • Computer Vision

Background:

  • Visual prostheses aim to restore sight to individuals with blindness.
  • Pixelized vision is a common model for simulating the output of visual prostheses.
  • Understanding the impact of visual display parameters is crucial for designing effective prostheses.

Purpose of the Study:

  • To evaluate a simulated pixelized prosthetic vision model using noncontiguous circular phosphenes.
  • To determine how phosphene and grid parameters affect facial recognition capabilities.

Main Methods:

  • A video headset presented faces through a pixelizing grid (10x10 to 32x32 dots).
  • Grid size, dot size, gap width, dot dropout rate, and gray-scale resolution were systematically varied.

Related Experiment Videos

  • Tests were conducted at high (99%) and low (12.5%) contrast levels.

    • Main Results:

    • Facial recognition accuracy was influenced by all tested stimulus parameters.
    • High-contrast recognition was accurate except with 70% dot dropout and two gray levels.
    • Low-contrast recognition remained significant even with adverse parameters like reduced grid area and wider gaps.
    • Learning and practice effects were observed, particularly under challenging conditions.

    Conclusions:

    • Reliable facial recognition is achievable with simplified pixelized grids.
    • These findings suggest the potential for functional visual prostheses that enable face recognition.