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

Graph matching vs mutual information maximization for object detection.

L B Sham1, M J Brady, S Schaal

  • 1California Institute of Technology, Computation and Neural Systems, Division for Biology, Pasadena 92215, USA. ladan@caltech.edu

Neural Networks : the Official Journal of the International Neural Network Society
|May 9, 2001
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

Fast Gate-Based Readout of Silicon Quantum Dots Using Josephson Parametric Amplification.

Physical review letters·2020
Same author

Automated gesture tracking in head-fixed mice.

Journal of neuroscience methods·2017
Same author

The anatomical and functional benefit of bevacizumab in the treatment of macular edema associated with Purtscher-like retinopathy.

Eye (London, England)·2014
Same author

Sleep fragmentation promotes NADPH oxidase 2-mediated adipose tissue inflammation leading to insulin resistance in mice.

International journal of obesity (2005)·2013
Same author

Thermal performance and limitations of silicon-substrate packaged GaAs laser arrays.

Applied optics·2010
Same author

Gallium arsenide laser-array-on-silicon package.

Applied optics·2010
Same journal

A practical design of backdoor trigger under frequency-based orthogonality constraints.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

EEG fine-grained visual semantic decoding via a multimodal framework.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Collaborative-adversarial jailbreaking: A propagation-aware attack framework for multi-agent code generation systems.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Theoretical analysis of the denoising autoencoder using Tweedie's formula.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

Frequency-based cross-attention fusion network for RGB-D salient object detection.

Neural networks : the official journal of the International Neural Network Society·2026
Same journal

HTNet: A self-supervised heterogeneous triple network for multi-modal data.

Neural networks : the official journal of the International Neural Network Society·2026
See all related articles

Labeled Graph Matching plus (LGM+) enhances visual pattern recognition, outperforming LGM and Mutual Information Maximization (MIM) on object detection tasks. LGM+ offers improved performance with simplicity and lower computational cost.

Area of Science:

  • Computer Vision
  • Pattern Recognition
  • Machine Learning

Background:

  • Labeled Graph Matching (LGM) is a foundational method in object vision and face recognition.
  • State-of-the-art methods include statistical approaches like Mutual Information Maximization (MIM).

Purpose of the Study:

  • To introduce LGM+, an extension of LGM for visual pattern recognition.
  • To compare the performance of LGM, LGM+, and an adapted MIM method on an object detection task.

Main Methods:

  • Implemented LGM and LGM+ algorithms.
  • Adapted the MIM method for multi-dimensional Gabor wavelet features.
  • Evaluated algorithms on a novel object detection dataset.

Main Results:

Related Experiment Videos

  • MIM with Gabor wavelets outperformed MIM on pixels and LGM.
  • LGM+ significantly surpassed LGM in performance.
  • LGM+ demonstrated superior results compared to the adapted MIM method.

    • Conclusions:

    • LGM+ represents a significant advancement over LGM for visual pattern recognition.
    • LGM+ retains LGM's advantages of simplicity and biological plausibility.
    • LGM+ offers a computationally efficient alternative to MIM-based methods.