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

Maximum confidence hidden markov modeling for face recognition.

Jen-Tzung Chien1, Chih-Pin Liao

  • 1Departmernt of Computer Science and Information Engineering, National Cheng Kung University, Tainan, Taiwan. jtchien@mail.ncku.edu.tw

IEEE Transactions on Pattern Analysis and Machine Intelligence
|February 16, 2008
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

Contrastive Mixture Diffusion Models.

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

Outlier-Aware Contrastive Learning.

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

Advancements in Clinical Evaluation and Regulatory Frameworks for AI-Driven Software as a Medical Device (SaMD).

IEEE open journal of engineering in medicine and biology·2024
Same author

Latent Semantic and Disentangled Attention.

IEEE transactions on pattern analysis and machine intelligence·2024
Same author

Methoxyhispolon Methyl Ether, a Hispolon Analog, Thwarts the SRC/STAT3/BCL-2 Axis to Provoke Human Triple-Negative Breast Cancer Cell Apoptosis In Vitro.

Biomedicines·2023
Same author

Hyperspectral signature-band extraction and learning: an example of sugar content prediction of Syzygium samarangense.

Scientific reports·2023
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
Same journal

Task-KV: Task-aware KV Cache Optimization via Semantic Differentiation of Attention Heads.

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

Achieving Text-based Person Retrieval with Any Granularity.

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

This study introduces Maximum Confidence Hidden Markov Modeling (MC-HMM) for robust 2D pattern recognition, enhancing face recognition accuracy with fewer features. The novel approach ensures compact and discriminative models for improved performance.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Pattern Recognition

Background:

  • Traditional Hidden Markov Models (HMMs) face challenges in achieving both model compactness and discriminability for complex pattern recognition tasks.
  • Existing discriminative training criteria may not sufficiently optimize for distinguishing between target and competing models, impacting recognition accuracy.

Purpose of the Study:

  • To develop a novel hybrid framework integrating feature extraction and Hidden Markov Modeling (HMM) for enhanced two-dimensional pattern recognition.
  • To introduce a new discriminative training criterion, Maximum Confidence Hidden Markov Modeling (MC-HMM), to improve model compactness and discriminability.
  • To apply the MC-HMM framework to face recognition, aiming for robust performance across variations in expressions and orientations.

Main Methods:

Related Experiment Videos

  • A hybrid framework combining feature extraction with HMM is proposed.
  • A new discriminative training criterion, derived from hypothesis testing theory, is developed to maximize confidence in target HMM state assignments.
  • The Maximum Confidence Hidden Markov Modeling (MC-HMM) incorporates a transformation matrix for discriminative facial feature extraction and uses closed-form solutions for continuous-density HMM parameters.
  • MC-HMM parameters are estimated and converged using the expectation-maximization procedure under the same confidence-based criterion.

Main Results:

  • The proposed MC-HMM method demonstrates robust segmentation capabilities, handling variations in facial expressions and orientations effectively.
  • Experimental results on the FERET and GTFD facial databases show superior performance compared to standard Maximum Likelihood and Minimum Classification Error HMMs.
  • The MC-HMM achieves higher recognition accuracies while utilizing lower feature dimensions, indicating improved efficiency and effectiveness.

Conclusions:

  • The developed MC-HMM framework offers a significant advancement in two-dimensional pattern recognition, particularly for face recognition tasks.
  • The novel discriminative training criterion effectively enhances model compactness and discriminability, leading to improved recognition performance.
  • MC-HMM provides a robust and accurate solution for face recognition, outperforming existing HMM-based approaches in terms of accuracy and feature dimensionality.