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: Dec 8, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
03:31

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

875

Towards Unsupervised Deep Image Enhancement with Generative Adversarial Network.

Zhangkai Ni, Wenhan Yang, Shiqi Wang

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |September 22, 2020
    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

    Magnetic Resonance Imaging and Computed Tomography of the Brain-50 Years of Innovation, With a Focus on the Future.

    Investigative radiology·2015
    Same author

    Magnetic resonance imaging and DWI features of orbital rhabdomyosarcoma.

    Eye science·2015
    Same author

    Large deletions in the NSDHL gene in two patients with CHILD syndrome.

    Acta dermato-venereologica·2015
    Same author

    Astrocytic tumour grading: a comparative study of three-dimensional pseudocontinuous arterial spin labelling, dynamic susceptibility contrast-enhanced perfusion-weighted imaging, and diffusion-weighted imaging.

    European radiology·2015
    Same author

    TGF-β1 expression is associated with invasion and metastasis of intrahepatic cholangiocarcinoma.

    Biological research·2015
    Same author

    Pharmacokinetic of 5 components after oral administration of Fructus Forsythiae by HPLC-MS/MS and the effects of harvest time and administration times.

    Journal of chromatography. B, Analytical technologies in the biomedical and life sciences·2015

    This study introduces an unsupervised image enhancement generative adversarial network (UEGAN) for improving photo aesthetics without paired data. The UEGAN effectively enhances image quality by learning desired characteristics in an unsupervised manner.

    Area of Science:

    • Computer Vision
    • Artificial Intelligence
    • Machine Learning

    Background:

    • Supervised learning methods for image enhancement require paired low-quality and expert-retouched images.
    • Expert retouching styles may not align with general user preferences.
    • A need exists for unsupervised methods to enhance image aesthetics based on desired characteristics.

    Purpose of the Study:

    • To present an unsupervised image enhancement generative adversarial network (UEGAN) for automatic photo enhancement.
    • To enable learning of image-to-image mapping without relying on large paired datasets.
    • To improve the aesthetic quality of images according to user-defined characteristics.

    Main Methods:

    • Developed an unsupervised image enhancement generative adversarial network (UEGAN).

    Related Experiment Videos

    Last Updated: Dec 8, 2025

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications
    03:31

    Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

    Published on: December 15, 2023

    875
  • Employed a single deep Generative Adversarial Network (GAN) with modulation and attention mechanisms.
  • Introduced two novel losses: fidelity loss (l2 regularization in VGG feature domain) and quality loss (relativistic hinge adversarial loss).

    • Main Results:

    • The UEGAN model effectively enhances image aesthetic quality.
    • Quantitative and qualitative results demonstrate the model's performance.
    • The unsupervised approach successfully learns desired image characteristics.

    Conclusions:

    • The proposed UEGAN offers an effective solution for unsupervised image enhancement.
    • The model can improve image aesthetics without paired training data.
    • This method provides a flexible alternative to supervised approaches for photo enhancement.