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

Person Reidentification Based on Elastic Projections.

Xuelong Li, Lina Liu, Xiaoqiang Lu

    IEEE Transactions on Neural Networks and Learning Systems
    |April 20, 2017
    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

    The Dual Effect of Selenium Application in Reducing Fusarium Wilt Disease Incidence in Banana and Producing Se-Enriched Fruits.

    Plants (Basel, Switzerland)·2024
    Same author

    New quantified climate reconstruction in southern China suggests a potential influence of winter cloud cover across East Asia during the Holocene.

    The Science of the total environment·2024
    Same author

    Discovery of 6-Fluoro-5-{4-[(5-fluoro-2-methyl-3-oxo-3,4-dihydroquinoxalin-6-yl)methyl]piperazin-1-yl}-<i>N</i>-methylpyridine-2-carboxamide (AZD9574): A CNS-Penetrant, PARP1-Selective Inhibitor.

    Journal of medicinal chemistry·2024
    Same author

    Stress responses, social support and death care experience among spouse caregivers of young and middle-aged patients with terminal cancer: a mixed-methods analysis.

    Frontiers in oncology·2024
    Same author

    Gap plasmonic properties of NPOM structures composed of gold nanoparticles and thin films.

    Scientific reports·2024
    Same author

    Prediction of prognosis and immune response in lung adenocarcinoma based on mitophagy and lactate-related gene signatures.

    International journal of clinical oncology·2024

    This study introduces elastic projections for person reidentification, enhancing accuracy by learning separate positive and negative projections. This method improves distinguishing individuals across different camera views.

    Area of Science:

    • Computer Vision
    • Machine Learning
    • Artificial Intelligence

    Background:

    • Person reidentification (re-ID) is crucial for multi-camera surveillance.
    • Existing methods often lose discriminative information by projecting features into a single subspace.
    • This can lead to poor performance in distinguishing individuals across different views.

    Purpose of the Study:

    • To propose a novel elastic projection method for person reidentification.
    • To learn a pairwise similarity measure that is both representative and discriminative.
    • To improve the accuracy and robustness of person re-ID systems.

    Main Methods:

    • Developed a method using elastic projections for person reidentification.
    • Introduced positive and negative projections to capture intrinsic attributes of same and different targets, respectively.

    Related Experiment Videos

  • Utilized prior information to guide the learning phase, enhancing discriminability.

    • Main Results:

    • The proposed elastic projection model demonstrated superior performance compared to existing methods.
    • Achieved better results on benchmark datasets like VIPeR, GRID, and CUHK.
    • Effectively reduced similarity between non-target individuals by making same targets more similar and different targets more distinct.

    Conclusions:

    • Elastic projections offer a more effective approach to person reidentification by preserving and enhancing discriminative features.
    • The method successfully bridges appearance variations while maintaining distinctiveness between individuals.
    • This work advances the state-of-the-art in non-overlapping multi-camera person reidentification.