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

Context quantization by kernel Fisher discriminant.

Mantao Xu1, Xiaolin Wu, Pasi Fränti

  • 1Department of Computer Science, University of Joensuu, 80101 Joensuu, Finland. xu@cs.joensuu.fi

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|January 27, 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

Music emotion classification based on random swap algorithm.

Scientific reports·2025
Same author

Advances and challenges of targeting epicardial adipose tissue (EAT) and perivascular adipose tissue (PVAT).

Cardiovascular diabetology·2025
Same author

CD22 TCR-engineered T cells exert antileukemia cytotoxicity without causing inflammatory responses.

Science advances·2025
Same author

Multi-omics analysis of copper metabolism-related molecular subtypes and risk stratification for osteosarcoma.

Discover oncology·2025
Same author

Pharmacokinetics of Metronidazole During Prolonged Intermittent Kidney Replacement Therapy.

Hemodialysis international. International Symposium on Home Hemodialysis·2025
Same author

Rapid Universal Detection of High-Risk and Low-Abundance Microbial Contaminations in CAR-T Cell Therapy.

Small methods·2025
Same journal

Change-Prior-Guided Unsupervised Change Detection of Heterogeneous Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

AgonicDreamer: Enhancing Multi-View Consistency in Text-to-3D Generation via Rectified Score Distillation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

BiCM-Prompt: Bidirectional Cross-Modal Prompt Tuning for Class-Incremental Learning on Multisource Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

GoP-based Quality Enhancement on Video Compression.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Align then Tensorize: Multi-Level Consistent Anchor Graph Learning for Scalable Multi-View Clustering.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Beyond Fidelity: Diverse Image Synthesis via Retrieval-Augmented Diffusion.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
See all related articles

New algorithms using multiclass and kernel Fisher discriminant improve context quantizer design for efficient conditional entropy coding. These methods offer practical implementation with simple scalar functions, outperforming previous approaches.

Area of Science:

  • Information Theory
  • Machine Learning
  • Data Compression

Background:

  • Optimal context quantizers aim to minimize conditional entropy for efficient coding.
  • A key challenge is the complexity of the quantizer mapping function in the context space.
  • Existing methods struggle with practical implementation due to complex mapping functions.

Purpose of the Study:

  • To develop novel algorithms for designing context quantizers in the context space.
  • To overcome the implementation difficulties associated with complex quantizer mapping functions.
  • To improve the performance of context quantization methods for conditional entropy coding.

Main Methods:

  • Proposed new algorithms based on multiclass Fisher discriminant and kernel Fisher discriminant (KFD).

Related Experiment Videos

  • Utilized KFD to project input context vectors onto a high-dimensional curve for better separability.
  • Compared performance against previous linear Fisher discriminant methods for context quantization.

    • Main Results:

    • The new algorithms, particularly KFD, demonstrated superior performance compared to the linear Fisher discriminant method.
    • The proposed methods achieved results close to the optimal minimum empirical conditional entropy quantizer.
    • Implemented a practical solution using a simple scalar quantizer mapping function, avoiding large lookup tables.

    Conclusions:

    • Multiclass and kernel Fisher discriminant provide effective approaches for designing practical context quantizers.
    • These methods significantly simplify implementation while maintaining high performance in conditional entropy coding.
    • The KFD approach offers a robust solution for linearly nonseparable quantizer cells.