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: May 13, 2026

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

Visual recognition algorithm for weakly labeled multi-source data fusion of remote sensing images.

Zemin Qiu1, Jiajun Zou1, Shaojiang Liu1

  • 1School of Information and Intelligence Engineering, Guangzhou Xinhua University, Dongguan, 523133, China.

Scientific Reports
|May 11, 2026
PubMed
Summary

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

Rb<sub>2</sub>HfCl<sub>6</sub>:Sb<sup>3+</sup>phosphors with tunable energy transfer for advanced information encryption and high-CRI WLEDS.

Nanotechnology·2026
Same author

Cross-modal semantic scene completion for complex urban traffic scenes with reachability geometric attention.

Scientific reports·2026
Same author

The impact of the primary surgeon's routine surgical experience on the clinical efficacy of robot-assisted knee arthroplasty.

Journal of robotic surgery·2026
Same author

Comparative study on the performance and early clinical outcomes of senior and junior orthopaedic surgeons in traditional versus robot-assisted total hip arthroplasty.

Journal of robotic surgery·2026
Same author

Multi-module collaborative 3D human body modeling algorithm based on PIFuHD.

Scientific reports·2026
Same author

Robot-assisted elimination of side disparity in acetabular cup placement by right-handed surgeons during total hip arthroplasty: a comparative study of manual and robotic techniques.

Journal of robotic surgery·2026

This study introduces a novel algorithm for remote sensing image recognition using weak annotations. It enhances fine-grained feature extraction and cross-domain generalization, achieving 94% accuracy.

Area of Science:

  • Computer Vision
  • Remote Sensing
  • Machine Learning

Background:

  • Existing fine-grained visual recognition methods struggle with strong annotation requirements and poor cross-domain generalization.
  • Remote sensing image recognition faces challenges due to limited labeled data and domain variability.

Purpose of the Study:

  • To develop a weakly annotated multi-source data fusion algorithm for improved remote sensing image recognition.
  • To enhance the utilization of weakly annotated data, enable precise fine-grained feature extraction, and boost cross-domain generalization.

Main Methods:

  • A self-supervised learning framework with attention mechanisms processes weak annotations, extracting features and filtering high-confidence samples.
  • A multimodal cross-attention fusion model integrates image and textual features into a shared space for fine-grained recognition.
Keywords:
Attention mechanismMulti-source dataVisual recognitionWeak labeling

Related Experiment Videos

Last Updated: May 13, 2026

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment
08:25

Combining Eye-tracking Data with an Analysis of Video Content from Free-viewing a Video of a Walk in an Urban Park Environment

Published on: May 7, 2019

  • Cross-domain contrastive loss is employed to identify domain-invariant features, improving generalization.

    • Main Results:

    • The proposed algorithm achieves 94% classification accuracy on traditional datasets, surpassing existing baseline models.
    • Demonstrates stable performance in real-world remote sensing scenarios.
    • Effectively utilizes weakly annotated data and enhances fine-grained feature extraction.

    Conclusions:

    • The weakly annotated multi-source data fusion algorithm offers a robust solution for remote sensing image recognition.
    • The method significantly improves cross-domain generalization capabilities, addressing key limitations of current approaches.
    • This work paves the way for more efficient and accurate remote sensing analysis with limited annotations.