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

Face recognition from a single training image under arbitrary unknown lighting using spherical harmonics.

Lei Zhang1, Dimitris Samaras

  • 1Computer Science Department, State University of New York, Stony Brook 11794-4400, USA. lzhang@cs.sunysb.edu.

IEEE Transactions on Pattern Analysis and Machine Intelligence
|March 11, 2006
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

Topology-Aware Segmentation Using Discrete Morse Theory.

... International Conference on Learning Representations·2026
Same author

TOPODIFFUSIONNET: A TOPOLOGY-AWARE DIFFUSION MODEL.

... International Conference on Learning Representations·2026
Same author

Unifying Top-down and Bottom-up Scanpath Prediction Using Transformers.

Proceedings. IEEE Computer Society Conference on Computer Vision and Pattern Recognition·2026
Same author

Look Hear: Gaze Prediction for Speech-directed Human Attention.

Computer vision - ECCV ... : ... European Conference on Computer Vision : proceedings. European Conference on Computer Vision·2026
Same author

Measuring and predicting where and when pathologists focus their visual attention while grading whole slide images of cancer.

Medical image analysis·2025
Same author

Label-Efficient Deep Color Deconvolution of Brightfield Multiplex IHC Images.

IEEE transactions on medical imaging·2025
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 two novel face recognition methods using spherical harmonics illumination representation. These techniques achieve high accuracy with minimal training data, even under challenging lighting conditions.

Area of Science:

  • Computer Vision
  • Image Processing
  • Machine Learning

Background:

  • Face recognition systems often struggle with varying illumination conditions.
  • Existing methods may require extensive training data or 3D shape information.
  • Spherical harmonics illumination representation offers a robust way to model lighting effects.

Purpose of the Study:

  • To develop novel face recognition methods robust to arbitrary unknown lighting.
  • To reduce the need for extensive training data and 3D shape information.
  • To accurately represent and recognize faces under diverse illumination scenarios.

Main Methods:

  • Two methods are proposed to estimate spherical harmonic basis images from a single training image.
  • Method 1 uses a statistical model based on 2D basis images for pose-invariant recognition.

Related Experiment Videos

  • Method 2 integrates spherical harmonics with a 3D morphable model for pose and illumination variation.

    • Main Results:

    • Both methods achieve high face recognition rates under various and multiple illumination conditions.
    • The proposed methods demonstrate comparable accuracy to systems with greater data requirements.
    • Experimental results validate the effectiveness of both proposed approaches.

    Conclusions:

    • Novel face recognition methods using spherical harmonics illumination representation are effective.
    • The techniques significantly reduce training data and 3D shape information requirements.
    • These methods offer a promising solution for robust face recognition in uncontrolled environments.