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

Real-time face detection and lip feature extraction using field-programmable gate arrays.

Duy Nguyen1, David Halupka, Parham Aarabi

  • 1Department of Electrical and Computer Engineering, University of Toronto, ON, Canada.

IEEE Transactions on Systems, Man, and Cybernetics. Part B, Cybernetics : a Publication of the IEEE Systems, Man, and Cybernetics Society
|August 15, 2006
PubMed
Summary

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

Rural physician deployment programs in five Southeast Asian countries: a policy and narrative review.

BMC health services research·2026
Same author

Field Monitoring of Microplastics in Urban Runoff: Treatment Efficiency of Biofilters.

Environmental science & technology·2026
Same author

First-in-Human Clinical Experience With Focal Pulsed Field and Radiofrequency Dual-Modality Ablation for Treatment Refractory Left Ventricular Summit PVCs.

Circulation. Arrhythmia and electrophysiology·2026
Same author

Allogeneic CD19 CAR T cells armed with an anti-rejection CD70 CAR overcome antigen escape and evade alloimmune responses.

Nature communications·2026
Same author

Exploring Civic Engagement and Well-Being Among Korean American Older Immigrants.

Journal of gerontological social work·2026
Same author

The Stain Reduction and Whitening Efficacy of an Enzyme-Enhanced Mouthwash: A Comparative Study.

Clinical, cosmetic and investigational dentistry·2026

This study introduces efficient face detection and lip feature extraction techniques using a naive Bayes classifier and edge representation. The real-time field-programmable gate array (FPGA) implementation offers significant size reduction and high accuracy for various lighting conditions.

Area of Science:

  • Computer Vision
  • Hardware Implementation
  • Machine Learning

Background:

  • Face detection and lip feature extraction are crucial for human-computer interaction and speech processing.
  • Existing methods often require substantial computational resources and model sizes.
  • Real-time processing on embedded systems remains a challenge.

Purpose of the Study:

  • To develop a novel, resource-efficient technique for face detection and lip feature extraction.
  • To implement these techniques on a real-time Field-Programmable Gate Array (FPGA) for practical applications.
  • To evaluate the performance and efficiency of the proposed system.

Main Methods:

  • Face detection utilizes a naive Bayes classifier on an edge-extracted image representation, significantly reducing model size.

Related Experiment Videos

  • Lip feature extraction focuses on contrast around the lip contour to determine mouth height and width.
  • A real-time FPGA system is designed to implement both proposed techniques.

    • Main Results:

    • The face detection model size is reduced to 5184 B, 2417 times smaller than comparable methods.
    • An 86.6% correct detection rate is achieved under diverse lighting conditions.
    • The FPGA system uses only 15050 logic cells, approximately six times less than comparable face detection systems.

    Conclusions:

    • The proposed naive Bayes classifier with edge representation offers a highly efficient face detection method.
    • The lip feature extraction technique provides valuable metrics for speech filtering applications.
    • The real-time FPGA implementation demonstrates the feasibility of deploying these advanced techniques on resource-constrained hardware.