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

Acquiring linear subspaces for face recognition under variable lighting.

Kuang-Chih Lee1, Jeffrey Ho, David J Kriegman

  • 1Beckman Institute and Computer Science Department, University of Illinois at Urbana-Champaign, Urbana, IL 61801, USA. klee10@uiuc.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|May 7, 2005
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

Experiences of Women Sexual Assault Survivors With Police in the Early Aftermath of Assault: Results From a Large-scale Prospective Study.

Journal of forensic nursing·2025
Same author

Rasch analysis of the new general self efficacy scale: an evaluation of its psychometric properties in older adults with low vision.

Health and quality of life outcomes·2024
Same author

Experiences of women sexual assault survivors with police in the early aftermath of assault: Results from a large-scale prospective study.

Research square·2024
Same author

Research with women sexual assault survivors presenting for emergency care is safe: Results from a multi-site, prospective observational cohort study.

Journal of psychiatric research·2024
Same author

Life space limitations in visually impaired older adults.

Optometry and vision science : official publication of the American Academy of Optometry·2024
Same author

Conducted Electrical Weapon Fire Risk in the Presence of Supplemental Oxygen.

Journal of special operations medicine : a peer reviewed journal for SOF medical professionals·2024
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

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

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

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

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

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

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

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

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

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

Learning Shape Anchors for Holistic Indoor Scene Understanding.

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

This study introduces a new method for face recognition using physical lighting. By capturing images under specific light configurations, researchers created effective low-dimensional linear spaces for recognition, avoiding complex 3D modeling.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Image Recognition

Background:

  • Image variations due to lighting are a challenge in object recognition, particularly for human faces.
  • Previous methods model these variations using low-dimensional linear spaces, often derived from extensive image sets or 3D models.
  • Existing techniques include Principal Component Analysis (PCA) on acquired images or synthetic renderings, and spherical harmonics for diffuse lighting.

Purpose of the Study:

  • To develop a method for constructing low-dimensional linear spaces directly from physical images of an object under controlled lighting.
  • To demonstrate that this subspace representation is effective for recognition under diverse lighting conditions.
  • To avoid complex intermediate steps like 3D reconstruction or the need for large training datasets.

Related Experiment Videos

Main Methods:

  • Configuring k point light sources (typically 5-9) in specific directions.
  • Acquiring k images of an object under these single-source lighting conditions.
  • Using these acquired images directly as basis vectors for a low-dimensional linear subspace.

Main Results:

  • The subspace generated from k physically lit images effectively represents object variations under a wide range of lighting.
  • This method yields subspaces comparable in effectiveness to those obtained through PCA on large datasets or synthetic rendering.
  • The approach successfully avoids the need for 3D reconstruction or complex diffuse lighting setups.

Conclusions:

  • Physical lighting configurations can directly generate effective basis images for low-dimensional linear spaces.
  • This technique offers a simplified and efficient alternative for creating robust representations for face recognition.
  • The method eliminates the need for extensive data acquisition or complex modeling, streamlining the recognition process.