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 14, 2026

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

Unsupervised object segmentation with a hybrid graph model (HGM).

Guangcan Liu1, Zhouchen Lin, Yong Yu

  • 1Department of Computer Science and Engineering, Shanghai Jiao Tong University, Shanghai, PR China. roth@sjtu.edu.cn

IEEE Transactions on Pattern Analysis and Machine Intelligence
|March 20, 2010
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

Diff-MEF: Cross-Modal Diffusion Framework With Text Prompts and Semantic Perception for Multi-Exposure Image Fusion.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same author

Artificial intelligence-assisted remote sensing observation, understanding, and decision.

Innovation (Cambridge (Mass.))·2026
Same author

URFusion: Unsupervised Unified Degradation-Robust Image Fusion Network.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2025
Same author

From human herpes virus-6 reactivation to autoimmune reactivity against tight junctions and neuronal antigens, to inflammation, depression, and chronic fatigue syndrome due to Long COVID.

Journal of medical virology·2024
Same author

Towards a major methodological shift in depression research by assessing continuous scores of recurrence of illness, lifetime and current suicidal behaviors and phenome features.

Journal of affective disorders·2024
Same author

Extraordinarily Time- and Memory-Efficient Large-Scale Canonical Correlation Analysis in Fourier Domain: From Shallow to Deep.

IEEE transactions on neural networks and learning systems·2023
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 hybrid graph model for unsupervised object segmentation, enabling automatic image analysis without labeled data. The method effectively combines sample relationships for accurate class-specific segmentation and recognition.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Artificial Intelligence

Background:

  • Unsupervised object segmentation is crucial for automated image analysis.
  • Existing methods often struggle to integrate class-specific information with local image features.
  • The need for effective methods that perform segmentation without annotated training data is significant.

Purpose of the Study:

  • To propose a novel hybrid graph model (HGM) for class-specific unsupervised object segmentation.
  • To develop a method that effectively utilizes both symmetric and asymmetric relationships among image samples.
  • To achieve simultaneous object segmentation and recognition within a unified framework.

Main Methods:

  • A hybrid graph model (HGM) is proposed, representing samples as vertices and relationships as directed/undirected edges.

Related Experiment Videos

Last Updated: Jun 14, 2026

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data
12:08

From Voxels to Knowledge: A Practical Guide to the Segmentation of Complex Electron Microscopy 3D-Data

Published on: August 13, 2014

  • Superpixels are used as vertices, with asymmetric relationships representing conditional dependence and symmetric relationships capturing color/texture similarity.
  • The model combines Markov chains from directed subgraphs and minimal cuts from undirected subgraphs to determine object boundaries.

    • Main Results:

    • The HGM successfully integrates top-down and bottom-up information for a unified segmentation and recognition process.
    • Experiments on 42 object classes (9,415 images) demonstrate promising results for unsupervised object segmentation.
    • The method shows effectiveness in handling class-specific information and local image similarities.

    Conclusions:

    • The proposed hybrid graph model offers an effective solution for class-specific unsupervised object segmentation.
    • The HGM facilitates simultaneous segmentation and recognition, advancing automated image analysis.
    • This approach provides a robust framework for object segmentation without the need for annotated training data.