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

Appearance-based face recognition and light-fields.

Ralph Gross1, Iain Matthews, Simon Baker

  • 1Robotics Institute, Carnegie Mellon University, Pittsburgh, PA 15213, USA. rgross@cs.smu.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|September 24, 2004
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

Optimising cancer support for rural residents from a refugee background: changing models of access and care.

Australian journal of primary health·2026
Same author

A description of patient experience after undergoing an emergency laparotomy.

World journal of emergency surgery : WJES·2026
Same author

Can China keep up its extraordinary research growth?

Nature·2026
Same author

ChIP happens: from biochemical origins to the modern omics toolbox for understanding steroid hormone receptors.

The Biochemical journal·2026
Same author

BioTriplex: a full-text annotated corpus for fine-tuning language models in gene-disease relation extraction tasks.

Bioinformatics (Oxford, England)·2026
Same author

Optimising the implementation of a universal web-based mental health service for Australian secondary schools: a cluster randomised controlled trial.

Child and adolescent psychiatry and mental health·2025
Same journal

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

HardFlow: Hard-Constrained Sampling for Flow-Matching Models Via Trajectory Optimization.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Industrial Brain: Self-Evolving Neuro-Symbolic Autonomy with Causal Resilience for Cyber-Physical Systems.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Adaptive Hardness-Driven Dictionary Distillation for Incomplete Streaming View Clustering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

Mixture of Global and Local Experts with Diffusion Transformer for Controllable Face Generation.

IEEE transactions on pattern analysis and machine intelligence·2026
See all related articles

This study introduces a new theory for object recognition using light-field data. The developed algorithm enhances face recognition across different poses by utilizing all available image pixels to estimate an eigen light-field.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Image Processing

Background:

  • Object recognition algorithms heavily rely on feature selection.
  • Appearance-based methods use pixel intensities, which represent light radiance.
  • The plenoptic function, or light-field, encompasses all radiance values.

Purpose of the Study:

  • To develop a theory for appearance-based object recognition using light-fields.
  • To create a robust face recognition algorithm that handles pose variations.
  • To leverage all available image data for improved recognition accuracy.

Main Methods:

  • Developing a theory of object recognition based on light-field properties.
  • Implementing an algorithm that utilizes all pixels from multiple images.

Related Experiment Videos

  • Estimating the object's eigen light-field as the primary feature set.

    • Main Results:

    • The proposed theory directly leads to a pose-invariant face recognition algorithm.
    • The algorithm effectively uses all available pixels, regardless of their source image.
    • Recognition is based on the estimated eigen light-field, analogous to pixel intensities.

    Conclusions:

    • The eigen light-field provides a powerful feature set for appearance-based recognition.
    • This approach offers a unified framework for object recognition across varying poses.
    • The method demonstrates potential for enhanced face recognition systems.