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

Sparse representation for coarse and fine object recognition.

Thang V Pham1, Arnold W M Smeulders

  • 1Intelligent Sensory Information Systems, Faculty of Science, University of Amsterdam, Kruislaan 403, 1098 SJ Amsterdam, The Netherlands. vietp@science.uva.nl

IEEE Transactions on Pattern Analysis and Machine Intelligence
|March 29, 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

HOOK2 downregulation compromises the tumorigenic and stemness properties of ovarian cancer cells by increasing endoplasmic reticulum stress.

Cell death & disease·2026
Same author

Discovery of cerebrospinal fluid biomarkers for different dementias using mass spectrometry-based proteomics.

Alzheimer's & dementia (Amsterdam, Netherlands)·2026
Same author

Phosphoproteomics of Hypertrophic Cardiomyopathy Patient Myocardium and Novel hiPSC-CM Model Reveal Protein Kinase A as a Modulator of Microtubule Repolymerization.

bioRxiv : the preprint server for biology·2026
Same author

Boosting the Speed and Accuracy of Protein Quantification Algorithms in Mass Spectrometry-Based Proteomics.

Journal of proteome research·2026
Same author

Cerebrospinal fluid proteomic analysis of long-term antihypertensive treatment with amlodipine and atenolol in hypertensive rats.

Scientific reports·2025
Same author

Proteome Differences in Smooth Muscle Cells from Diabetic and Non-Diabetic Abdominal Aortic Aneurysm Patients Reveal Metformin-Induced Mechanisms.

Medical sciences (Basel, Switzerland)·2025
Same journal

TraGraph-GS: Trajectory Graph-based Gaussian Splatting for Arbitrary Large-Scale Scene Rendering.

IEEE transactions on pattern analysis and machine intelligence·2026
Same journal

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving.

IEEE transactions on pattern analysis and machine intelligence·2026
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
See all related articles

This study introduces a novel sparse object representation using Gaussian differential bases. This method achieves real-time object recognition with high accuracy, outperforming PCA in efficiency and flexibility.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Image Processing

Background:

  • Object recognition is a fundamental task in computer vision.
  • Existing methods like PCA have limitations in terms of data storage, retraining, and real-time performance during object search.

Purpose of the Study:

  • To develop a sparse, multiscale object representation for efficient and accurate recognition.
  • To overcome the limitations of traditional methods like PCA in object recognition tasks.

Main Methods:

  • Utilizing a large dictionary of Gaussian differential basis functions for object appearance.
  • Employing the matching pursuit algorithm for learning the representation.
  • Implementing polynomial approximation for object recognition, converting image matching to polynomial evaluation.

Related Experiment Videos

Main Results:

  • The proposed method achieves high recognition accuracy on the COIL-100 dataset.
  • Demonstrated real-time performance, significantly outperforming PCA in scenarios requiring multi-location image scanning.
  • The representation allows for easy addition of new objects without retraining existing models.

Conclusions:

  • The sparse, multiscale representation offers a flexible and efficient alternative for object recognition.
  • The method is suitable for both coarse and fine object recognition, including pose estimation.
  • This approach provides significant advantages over PCA for real-world object recognition applications.