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 Video

Updated: Jun 26, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

Natural image statistics and low-complexity feature selection.

Manuela Vasconcelos1, Nuno Vasconcelos

  • 1Statistical Visual Computing Laboratory, UCSD, La Jolla, CA 92093, USA. maspcv@gmail.com

IEEE Transactions on Pattern Analysis and Machine Intelligence
|December 27, 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

Rapid Artificial Intelligence Autoplanning Rivals Manual Expert Planning for Cervical Brachytherapy.

Practical radiation oncology·2026
Same author

Macula Spatial Patterns and Their Association With Central Visual Field Progression in Glaucoma Using Artificial Intelligence.

Journal of glaucoma·2026
Same author

Integration of single-click, AI-based brachytherapy auto-planning for cervical cancer within a treatment planning system.

Brachytherapy·2025
Same author

Deep Learning Estimation of 24-2 Visual Field Map From Optic Nerve Head Optical Coherence Tomography Angiography.

Journal of glaucoma·2025
Same author

Neural network dose prediction for cervical brachytherapy: Overcoming data scarcity for applicator-specific models.

Medical physics·2024
Same author

Deep Learning Estimation of 10-2 Visual Field Map Based on Macular Optical Coherence Tomography Angiography Measurements.

American journal of ophthalmology·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

Low-complexity feature selection for visual recognition is feasible, as high-order feature dependencies offer minimal information for image discrimination. New algorithms outperform existing methods, confirming low complexity is key for effective natural image classification.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Pattern Recognition

Background:

  • Feature selection is crucial for efficient visual recognition.
  • High-order feature dependencies in natural images may not significantly improve classification.
  • Existing methods for feature selection can be computationally intensive.

Purpose of the Study:

  • To analyze low-complexity feature selection in visual recognition.
  • To formally characterize the hypothesis that high-order feature dependencies contain little discriminative information.
  • To develop and evaluate new feature selection algorithms.

Main Methods:

  • Introduction of concepts: conjunctive interference and decomposability order.
  • Derivation of necessary and sufficient conditions for low-complexity feature selection.

Related Experiment Videos

Last Updated: Jun 26, 2026

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances
07:35

Selecting Multiple Biomarker Subsets with Similarly Effective Binary Classification Performances

Published on: October 11, 2018

  • Development of feature selection algorithms for varying complexity levels.
  • Comparison of new algorithms with existing information-theoretic methods.

    • Main Results:

    • Intrinsic complexity of feature selection depends on decomposability order, not dimension.
    • New algorithms outperform existing information-theoretic methods.
    • For image classification, modeling feature dependencies shows diminishing returns, with decomposability order 1 yielding optimal results.
    • A generic law for bandpass features in natural images is proposed.

    Conclusions:

    • Low-complexity feature selection is achievable and effective for visual recognition.
    • The decomposability order is a critical factor in feature selection complexity.
    • Feature dependencies have limited additional benefit beyond a certain complexity level in natural image classification.