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

Computational and performance aspects of PCA-based face-recognition algorithms.

H Moon1, P J Phillips

  • 1Department of Electrical and Computer Engineering, State University of New York at Buffalo, Amherst, NY 14260, USA. moon@acsu.buffalo.edu

Perception
|May 26, 2001
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

Cross-architecture tuning of silicon and SiGe-based quantum devices using machine learning.

Scientific reports·2024
Same author

Search for a Dark Photon and an Invisible Dark Higgs Boson in μ^{+}μ^{-} and Missing Energy Final States with the Belle II Experiment.

Physical review letters·2023
Same author

Measurement of the Λ_{c}^{+} Lifetime.

Physical review letters·2023
Same author

Semimetal to semiconductor transition in Bi/TiO<sub>2</sub> core/shell nanowires.

Nanoscale advances·2022
Same author

Precise Measurement of the D^{0} and D^{+} Lifetimes at Belle II.

Physical review letters·2021
Same author

Machine learning enables completely automatic tuning of a quantum device faster than human experts.

Nature communications·2020
Same journal

Benchmarking spatial discrimination thresholds of two-frame motion defined forms compared to luminance and stereoscopic defined forms.

Perception·2026
Same journal

The effect of face masks on the perception of trustworthiness and competence in individuals with autistic traits.

Perception·2026
Same journal

The importance of external features for categorizing ethnicity: can Koreans identify Korean, Japanese, and Chinese faces?

Perception·2026
Same journal

Interoception, alexithymia, and motor congruency: Psychological drivers of body ownership in virtual reality.

Perception·2026
Same journal

Emotional valence, not arousal, influences taste perception.

Perception·2026
Same journal

Consistency and changes in eye movements during static, dynamic, and audio-visual emotion processing.

Perception·2026
See all related articles

The choice of similarity measure significantly impacts face recognition algorithm performance. A +/- 10% performance difference is crucial for distinguishing between principal component analysis (PCA) face recognition algorithms.

Area of Science:

  • Computer Science
  • Psychology
  • Biometrics

Background:

  • Principal Component Analysis (PCA) is foundational to many face recognition studies.
  • Implementing PCA for face recognition involves critical design choices.
  • A generic modular PCA algorithm framework is introduced to analyze these decisions.

Purpose of the Study:

  • To investigate the impact of various design decisions on PCA-based face recognition algorithms.
  • To evaluate different implementations of PCA algorithm modules.
  • To analyze undocumented design choices in face recognition literature.

Main Methods:

  • Utilized the September 1996 FERET evaluation protocol for algorithm assessment.
  • Experimented with variations in illumination normalization.

Related Experiment Videos

  • Assessed the effects of JPEG and wavelet image compression on performance.
  • Varied the number of eigenvectors and the similarity measure for classification.

    • Main Results:

    • The similarity measure demonstrated the most substantial impact on algorithm performance.
    • Performance differences of +/- 10% are necessary to differentiate between algorithms.
    • Algorithm performance variability was examined across different facial image sets.

    Conclusions:

    • The selection of the similarity measure is paramount for optimizing PCA face recognition systems.
    • Robust evaluation requires distinguishing algorithms by at least a 10% performance margin.
    • Understanding design choices is key to advancing face recognition technology.