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

A unified framework for image retrieval using keyword and visual features.

Feng Jing1, Mingling Li, Hong-Jiang Zhang

  • 1Computer Science Department, Tsinghua University, Beijing 100084, China. jingfeng00@mails.tsinghua.edu.cn

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|July 21, 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

CK2 phosphorylates TNFAIP1 to affect its subcellular localization and interaction with PCNA.

Molecular biology reports·2009
Same author

Reamed and unreamed intramedullary nailing for the treatment of open and closed tibial fractures: a subgroup analysis of randomised trials.

International orthopaedics·2009
Same author

Selective COX-2 inhibitor versus nonselective COX-1 and COX-2 inhibitor in the prevention of heterotopic ossification after total hip arthroplasty: a meta-analysis of randomised trials.

International orthopaedics·2009
Same author

[Study on evaluating sex determining region of the Y as an engrafting track of BMSCs transplantation for repairing osteonecrosis of the femoral head of rabbit].

Zhongguo xiu fu chong jian wai ke za zhi = Zhongguo xiufu chongjian waike zazhi = Chinese journal of reparative and reconstructive surgery·2009
Same author

Positive association between benign familial infantile convulsions and LGI4.

Brain & development·2009
Same author

Catalytic enantioselective synthesis of chiral phthalides by efficient reductive cyclization of 2-acylarylcarboxylates under aqueous transfer hydrogenation conditions.

Organic letters·2009
Same journal

Style-Aware Contrastive Test-Time Adaptation: A Dual-Cache Model for Robust Vision-Language Alignment.

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

Semantic Frame Interpolation.

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

Physics-Guided Cross-Modal Decoupling with Test-Time Adaptation for Hyperspectral Image Restoration.

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

Change-Prior-Guided Unsupervised Change Detection of Heterogeneous Remote Sensing Images.

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

AgonicDreamer: Enhancing Multi-View Consistency in Text-to-3D Generation via Rectified Score Distillation.

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

BiCM-Prompt: Bidirectional Cross-Modal Prompt Tuning for Class-Incremental Learning on Multisource Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
See all related articles

This study introduces a unified image retrieval framework combining keywords and visual features. It enhances search accuracy and efficiency through novel relevance feedback and keyword propagation methods.

Area of Science:

  • Computer Science
  • Information Retrieval
  • Machine Learning

Background:

  • Traditional image retrieval systems often struggle to balance keyword-based and visual feature-based searches.
  • Existing methods may lack efficiency in handling large datasets and incorporating user feedback effectively.

Purpose of the Study:

  • To propose a unified image retrieval framework integrating both textual (keyword) and visual information.
  • To enhance the accuracy and efficiency of image retrieval through advanced relevance feedback mechanisms.
  • To develop methods for semantic concept representation and keyword propagation in unlabeled image datasets.

Main Methods:

  • Building statistical models from labeled images to represent semantic concepts and propagate keywords to unlabeled images.

Related Experiment Videos

  • Implementing periodic model updates using user-provided relevance feedback for knowledge accumulation.
  • Developing efficient similarity measures and relevance feedback schemes for both keyword and image example queries.
  • Introducing an entropy-based active learning strategy for improved keyword-based relevance feedback.
  • Proposing a new algorithm for estimating keyword features in image example queries.

    • Main Results:

    • The proposed framework effectively combines keyword models with visual features for improved retrieval.
    • The entropy-based active learning strategy enhances the efficiency of relevance feedback in keyword queries.
    • The new algorithm for keyword feature estimation in image example queries outperforms existing methods.
    • Experimental results validate the effectiveness and efficiency of the unified image retrieval framework.

    Conclusions:

    • The unified framework offers a robust solution for image retrieval by leveraging both semantic and visual data.
    • The integration of relevance feedback and active learning significantly improves retrieval performance and user experience.
    • The proposed methods represent a notable advancement in the field of content-based image retrieval.