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: Jul 9, 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

559

An Efficient Single Image De-Raining Model With Decoupled Deep Networks.

Wencheng Li, Gang Chen, Yi Chang

    IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
    |November 30, 2023
    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 local control of radiotherapy following Ivor-Lewis esophagectomy in the patients with stage II A middle-third thoracic esophageal cancer].

    Zhonghua wai ke za zhi [Chinese journal of surgery]·2008
    Same author

    [Construction of GSK-3beta-targeting RNAi adenovirus vector and the influence of Wnt/beta-catenin pathway in proliferation of human thyrocytes].

    Zhonghua yi xue za zhi·2008
    Same author

    Carbon nanotube/polystyrene composite electrode for microchip electrophoretic determination of rutin and quercetin in Flos Sophorae Immaturus.

    Talanta·2008
    Same author

    [The effects and therapeutic duration of oral corticosteroids in patients with acute exacerbation of chronic obstructive pulmonary diseases].

    Zhonghua jie he he hu xi za zhi = Zhonghua jiehe he huxi zazhi = Chinese journal of tuberculosis and respiratory diseases·2008
    Same author

    Magnetic loading of carbon nanotube/nano-Fe(3)O(4) composite for electrochemical sensing.

    Talanta·2008
    Same author

    Isoliquiritigenin, a flavonoid from licorice, plays a dual role in regulating gastrointestinal motility in vitro and in vivo.

    Phytotherapy research : PTR·2008
    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
    Same journal

    GoP-based Quality Enhancement on Video Compression.

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

    Align Then Tensorize: Multi-Level Consistent Anchor Graph Learning for Scalable Multi-View Clustering.

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

    Beyond Fidelity: Diverse Image Synthesis via Retrieval-Augmented Diffusion.

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

    This study introduces a new deep learning (DL) model, DLINet, for single image de-raining. DLINet effectively separates rain detection and intensity estimation, improving de-raining performance on computer vision tasks.

    Area of Science:

    • Computer Vision
    • Deep Learning
    • Image Restoration

    Background:

    • Rain streaks severely degrade image visibility in outdoor computer vision applications.
    • Deep learning (DL) has advanced de-raining methods, but homogeneous architectures struggle with distinct rain characteristics.
    • Existing methods often overlook the discrepancy between rain location detection and intensity estimation, leading to performance degradation.

    Purpose of the Study:

    • To propose a novel heterogeneous de-raining architecture, DLINet, that decouples rain location detection and rain intensity estimation.
    • To address feature interference and representation degradation issues in current DL-based de-raining methods.
    • To enhance de-raining performance for computer vision applications.

    Main Methods:

    More Related Videos

    Deep Neural Networks for Image-Based Dietary Assessment
    13:19

    Deep Neural Networks for Image-Based Dietary Assessment

    Published on: March 13, 2021

    9.2K
    Deep Learning-Based Segmentation of Cryo-Electron Tomograms
    10:25

    Deep Learning-Based Segmentation of Cryo-Electron Tomograms

    Published on: November 11, 2022

    8.8K

    Related Experiment Videos

    Last Updated: Jul 9, 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

    559
    Deep Neural Networks for Image-Based Dietary Assessment
    13:19

    Deep Neural Networks for Image-Based Dietary Assessment

    Published on: March 13, 2021

    9.2K
    Deep Learning-Based Segmentation of Cryo-Electron Tomograms
    10:25

    Deep Learning-Based Segmentation of Cryo-Electron Tomograms

    Published on: November 11, 2022

    8.8K
  • Developed a heterogeneous de-raining architecture (DLINet) with dedicated subnetworks for rain location detection and rain intensity estimation.
  • Implemented a high-order collaborative network to manage dynamic inter-layer interactions between the decoupled subnetworks.
  • Introduced a novel training strategy with task-oriented supervision using labels from joint training.

    • Main Results:

    • DLINet demonstrates superior performance compared to existing state-of-the-art de-raining methods.
    • The decoupled approach effectively mitigates feature interference and improves representation ability.
    • Experiments on synthetic and real-world datasets validate the method's advantages.

    Conclusions:

    • The proposed DLINet architecture offers a significant advancement in single image de-raining.
    • Decoupling rain location detection and intensity estimation with dedicated subnetworks is crucial for optimal performance.
    • The novel training strategy and collaborative network enhance the effectiveness of de-raining.