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

Indexing hierarchical structures using graph spectra.

Ali Shokoufandeh1, Diego Macrini, Sven Dickinson

  • 1Department of Computer Science, College of Enegineering, 3141 Chestnut St., Philadelphia, PA 19104, USA. ashokouf@cs.drexel.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence
|July 15, 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

How 'Neural' is a Neural Foundation Model?

ArXiv·2026
Same author

Automated diagnosis of usual interstitial pneumonia on chest CT via the mean curvature of isophotes.

European journal of radiology open·2026
Same author

Probabilistic Directed Distance Fields for Ray-Based Shape Representations.

IEEE transactions on pattern analysis and machine intelligence·2025
Same author

Modular, focused data science education improves biomedical learners' abilities: A study of the Data and Analytics for Research Training (DART) program.

PLoS computational biology·2025
Same author

Orientation fields predict human perception of 3D shape from shading.

Proceedings of the National Academy of Sciences of the United States of America·2025
Same author

Functional organization and natural scene responses across mouse visual cortical areas revealed with encoding manifolds.

bioRxiv : the preprint server for biology·2024
Same journal

Relation DETR+: Exploring Explicit Position Relation Prior for Dense Prediction.

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

RBF++: Quantifying and Optimizing Reasoning Boundaries across Measurable and Unmeasurable Capabilities for Chain-of-Thought Reasoning.

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

CAFE: Cross-View Adaptive Fusion and Cluster Center Enhancement for Robust Multi-View Clustering.

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

DIVER: Reinforced Diffusion Breaks Imitation Bottlenecks in End-to-End Autonomous Driving.

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

Ethics-Aware Safe Reinforcement Learning for Rare-Event Risk Control in Interactive Urban Driving.

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

Learning Shape Anchors for Holistic Indoor Scene Understanding.

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

This study introduces a novel framework for indexing hierarchical image structures using directed acyclic graphs (DAGs). The topological signature efficiently retrieves 3D object recognition candidates, even with occlusions.

Area of Science:

  • Computer Vision
  • Image Analysis
  • Machine Learning

Background:

  • Hierarchical image structures are crucial for representing part relationships, scale, and multiresolution features in computer vision.
  • Existing methods face challenges in efficiently indexing and retrieving complex hierarchical image data.

Purpose of the Study:

  • To develop a robust framework for indexing hierarchical image structures represented as directed acyclic graphs (DAGs).
  • To enable efficient retrieval of 3D object recognition candidates from large databases.

Main Methods:

  • Embedding the topological structure of DAGs into a low-dimensional vector space using a novel spectral characterization.
  • Employing a nearest-neighbor search for efficient candidate retrieval.
  • Developing a voting mechanism within DAG subspaces to handle large-scale occlusions.

Related Experiment Videos

Main Results:

  • The proposed topological signature is insensitive to minor perturbations in graph structure (noise, occlusion, node changes).
  • The voting mechanism effectively accumulates local evidence for robust recognition under occlusion.
  • Demonstrated successful indexing and retrieval in view-based 3D object recognition using shock graphs.

Conclusions:

  • The developed framework provides an efficient and robust method for indexing hierarchical image representations.
  • The topological signature and voting strategy significantly improve 3D object recognition performance, especially in the presence of occlusions.