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

Generic Learning-Based Ensemble Framework for Small Sample Size Face Recognition in Multi-Camera Networks.

Cuicui Zhang1, Xuefeng Liang2, Takashi Matsuyama3

  • 1Graduate School of Informatices, Kyoto University, Kyoto 606-8501, Japan. zhang@vision.kuee.kyoto-u.ac.jp.

Sensors (Basel, Switzerland)
|December 11, 2014
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

Spatial accessibility analysis of vaccination units in Gansu Province, China (2025).

Human vaccines & immunotherapeutics·2026
Same author

Assessing spatial accessibility of rabies post-exposure prophylaxis clinics in Gansu Province, China.

Frontiers in public health·2026
Same author

The U-shaped relationship between triglyceride glucose-body mass index and prognosis of sepsis patients: a retrospective study.

BMC infectious diseases·2026
Same author

Dual-layer orthogonal acousto-optic sensor for alias-suppressed high-precision DOA sensing.

Optics express·2026
Same author

Effects and mechanisms of different organic manure on minimizing cadmium concentrations in rice grains.

International journal of phytoremediation·2026
Same author

Flavokawain A Mitigates Mycoplasma pneumoniae-Induced Pneumonia by Inhibiting ERK/JNK/NF-κB Pathway in a Mouse Model.

Journal of biochemical and molecular toxicology·2026

This study introduces a new ensemble framework to improve person re-identification in multi-camera systems. It effectively addresses the small sample size problem in face recognition, enhancing surveillance accuracy.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Artificial Intelligence

Background:

  • Multi-camera networks are crucial for surveillance, but person re-identification faces challenges.
  • Face recognition for re-identification suffers from the small sample size (SSS) problem due to limited training data and unconstrained conditions.
  • Existing ensemble methods struggle with generating diverse base classifiers and managing the diversity-accuracy trade-off.

Purpose of the Study:

  • To propose a novel generic learning-based ensemble framework to overcome the SSS problem in person re-identification.
  • To enhance the generation of diverse base classifiers from limited data.
  • To mitigate the diversity/accuracy dilemma in ensemble learning.

Main Methods:

  • Augmenting small datasets by generating new samples based on a generic distribution.

Related Experiment Videos

  • Employing a tailored 0-1 knapsack algorithm to select diverse and accurate base classifiers for the ensemble.
  • Developing a generic learning-based ensemble framework for face recognition in multi-camera networks.

    • Main Results:

    • The proposed framework successfully generates more diverse base classifiers from an expanded face space.
    • The tailored knapsack algorithm effectively selects appropriate base classifiers, alleviating the diversity/accuracy dilemma.
    • Experimental results on four benchmarks demonstrate superior performance in addressing the SSS problem compared to state-of-the-art methods.

    Conclusions:

    • The novel ensemble framework provides a robust solution for person re-identification under the small sample size constraint.
    • The data augmentation and classifier selection strategies significantly improve the effectiveness of ensemble methods in challenging scenarios.
    • This approach enhances the reliability and accuracy of video-based surveillance systems.