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

Evolutionary optimization of a hierarchical object recognition model.

Georg Schneider1, Heiko Wersing, Bernhard Sendhoff

  • 1Audi Electronics Venture GmbH, D-85045 Ingolstadt, Germany. georg2.schneider@audi.de

IEEE Transactions on Systems, Man, and Cybernetics. Part B, Cybernetics : a Publication of the IEEE Systems, Man, and Cybernetics Society
|June 24, 2005
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

Double-trap measurement of the proton magnetic moment at 0.3 parts per billion precision.

Science (New York, N.Y.)·2017
Same author

Test Problems for Large-Scale Multiobjective and Many-Objective Optimization.

IEEE transactions on cybernetics·2017
Same author

Toward Self-Referential Autonomous Learning of Object and Situation Models.

Cognitive computation·2016
Same author

Structural synaptic plasticity has high memory capacity and can explain graded amnesia, catastrophic forgetting, and the spacing effect.

PloS one·2014
Same author

Powerful sequence similarity search methods and in-depth manual analyses can identify remote homologs in many apparently "orphan" viral proteins.

Journal of virology·2013
Same author

Tmem79/Matt is the matted mouse gene and is a predisposing gene for atopic dermatitis in human subjects.

The Journal of allergy and clinical immunology·2013
Same journal

Strategic Ability Updating in Concurrent Games by Coalitional Commitment.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2015
Same journal

Meta-Analysis of the First Facial Expression Recognition Challenge.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
Same journal

Adjustable model-based fusion method for multispectral and panchromatic images.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
Same journal

Face Feature Weighted Fusion Based on Fuzzy Membership Degree for Video Face Recognition.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
Same journal

A New Adaptive Fast Cellular Automaton Neighborhood Detection and Rule Identification Algorithm.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
Same journal

Human-arm-and-hand-dynamic model with variability analyses for a stylus-based haptic interface.

IEEE transactions on systems, man, and cybernetics. Part B, Cybernetics : a publication of the IEEE Systems, Man, and Cybernetics Society·2012
See all related articles

This study optimizes artificial neural network architectures for 3-D object recognition using biologically inspired models and evolutionary algorithms. Direct feature coding improves generalization on known data, while both direct and indirect methods achieve competitive results on new datasets.

Area of Science:

  • Computer Vision
  • Artificial Intelligence
  • Computational Neuroscience

Background:

  • Designing artificial neural network architectures, especially for 3-D object recognition, is challenging due to high-dimensional search spaces.
  • Standard optimization algorithms often yield suboptimal network structures for complex vision tasks.

Purpose of the Study:

  • To address the challenge of network architecture design for 3-D object classification.
  • To improve the optimization of features and nonlinearities in hierarchical vision networks.
  • To compare direct and indirect coding methods for evolutionary optimization of network features.

Main Methods:

  • Utilized biologically inspired hierarchical vision models to reduce search space dimensionality.
  • Employed evolutionary optimization techniques to determine optimal features and nonlinearities.

Related Experiment Videos

  • Compared direct feature coding with indirect coding via unsupervised learning embedded in evolution.
  • Fitness evaluated using nearest-neighbor classification on benchmark datasets, assessing first- and second-order generalization.

    • Main Results:

    • Direct feature coding resulted in superior first-order generalization (classification of unseen views from the same dataset).
    • Both direct and indirect coding methods demonstrated comparable second-order generalization (classification on a separate dataset).
    • Optimized networks achieved highly competitive performance against state-of-the-art recognition systems.

    Conclusions:

    • Biologically inspired models and evolutionary optimization effectively tackle complex neural network architecture design.
    • Indirect coding offers a viable alternative to direct coding, particularly for cross-dataset generalization.
    • The developed methods yield robust and competitive 3-D object recognition systems.