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: Jan 16, 2026

Author Spotlight: Enhancing Skin Model Diversity with Cost-Effective 3D Cellular Models
08:32

Author Spotlight: Enhancing Skin Model Diversity with Cost-Effective 3D Cellular Models

Published on: October 20, 2023

3.9K

Source-Free Model Adaptation for Unsupervised 3D Object Retrieval.

Dan Song, Yiyao Wu, Yuting Ling

    IEEE Transactions on Visualization and Computer Graphics
    |October 2, 2025
    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

    Barriers and Facilitators of Physical Activity in People Living With HIV: A Systematic Review of Qualitative Studies.

    Journal of the International Association of Providers of AIDS Care·2024
    Same author

    LC-MS/MS-based bioanalysis of branched-chain and aromatic amino acids in human serum.

    Bioanalysis·2024
    Same author

    Tourmaline triggered biofilm transformation: Boosting ultrafiltration efficiency and fouling resistance.

    Water research·2024
    Same author

    Study of long-term effects of pelvic radiotherapy on the function of bone marrow in recurrent cervical cancer patients.

    International journal of medical sciences·2024
    Same author

    Occurrence and emission characteristics of microplastics in agricultural surface runoff under different natural rainfall and short-term fertilizer application.

    Journal of hazardous materials·2024
    Same author

    Lin28a forms an RNA-binding complex with Igf2bp3 to regulate m<sup>6</sup>A-modified stress response genes in stress granules of muscle stem cells.

    Cell proliferation·2024
    Same journal

    MesoSplats: Texture Synthesis with Gaussian Splatting.

    IEEE transactions on visualization and computer graphics·2026
    Same journal

    GLLA: A Unified Force-Directed Graph Layout Framework Supporting Local Adjustments.

    IEEE transactions on visualization and computer graphics·2026
    Same journal

    Multi-Perception Crowd: Learning to combine entity and implicit perception for diverse crowd simulation.

    IEEE transactions on visualization and computer graphics·2026
    Same journal

    Hiding in Plain Sight: Camouflaging Real-world Objects.

    IEEE transactions on visualization and computer graphics·2026
    Same journal

    RTF2Mesh: Restricted Tangent Face Based Mesh Compression With Neural Displacement Fields.

    IEEE transactions on visualization and computer graphics·2026
    Same journal

    Practical Occluder Generation for Mobile Games.

    IEEE transactions on visualization and computer graphics·2026
    See all related articles

    This study introduces source-free model adaptation for unsupervised 3D object retrieval, overcoming data access limitations. The method effectively adapts models without source data, improving 3D object management performance.

    Area of Science:

    • Computer Vision
    • Machine Learning
    • 3D Data Analysis

    Background:

    • Unsupervised 3D object retrieval faces challenges due to high annotation costs and limited data accessibility.
    • Existing transfer learning methods require access to source data and labels, which are often restricted.
    • Privacy concerns, computational limits, and other restrictions hinder the use of labeled resources.

    Purpose of the Study:

    • To propose a novel source-free model adaptation approach for unsupervised 3D object management.
    • To enable effective 3D object retrieval without direct access to the original source data or labels.
    • To address the limitations of traditional transfer learning in scenarios with restricted data access.

    Main Methods:

    • Developed a source-free model adaptation task utilizing a pre-trained model.

    Related Experiment Videos

    Last Updated: Jan 16, 2026

    Author Spotlight: Enhancing Skin Model Diversity with Cost-Effective 3D Cellular Models
    08:32

    Author Spotlight: Enhancing Skin Model Diversity with Cost-Effective 3D Cellular Models

    Published on: October 20, 2023

    3.9K
  • Computed representative prototypes to approximate the source feature distribution.
  • Designed a bidirectional cumulative confidence-based adaptation strategy for aligning unlabeled data.
  • Implemented a dual-model distillation mechanism to generate source hypotheses in the absence of ground-truth labels.

    • Main Results:

    • Demonstrated the superiority of the proposed method on cross-domain (NTU-PSB) and cross-modality (MI3DOR) retrieval benchmarks.
    • Achieved significant performance improvements in unsupervised 3D object retrieval.
    • Validated the effectiveness of the approach even without access to raw source data.

    Conclusions:

    • The proposed source-free model adaptation method effectively enhances unsupervised 3D object retrieval performance.
    • The technique successfully overcomes the limitations of data accessibility and annotation costs.
    • This approach offers a viable solution for 3D object management in resource-constrained environments.