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 13, 2026

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

1.0K

Source-Free Domain-Adaptive Semi-Supervised Learning for Object Detection in CCTV Images.

Hyejin Shin1, Gye-Young Kim1

  • 1Department of AI·SW Convergence, Soongsil University, Seoul 06978, Republic of Korea.

Sensors (Basel, Switzerland)
|January 10, 2026
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

WEDGE-Net: Wavelet-Driven Memory-Efficient Anomaly Detection for Industrial Edge Computing.

Sensors (Basel, Switzerland)·2026
Same author

Multidirectional Trunk Movements Reveal Hidden Symmetry Loss in Stroke: An Electromyography-Based Comparative Study.

Medicina (Kaunas, Lithuania)·2025
Same author

Highly accurate image registration for 3D multiplexed cyclic imaging using dense labeling in expandable tissue gels.

PLoS biology·2025
Same author

The Effects of Upper Extremity Closed Kinetic Chain Exercise Combined with Biofeedback in Stroke Survivors: A Pilot Study.

NeuroRehabilitation·2025
Same author

Doubling multiplexed imaging capability via spatial expression pattern-guided protein pairing and computational unmixing.

Communications biology·2025
Same author

Deep Supervised Attention Network for Dynamic Scene Deblurring.

Sensors (Basel, Switzerland)·2025
Same journal

RETRACTED: Zhang et al. A Novel Framework for Reconstruction and Imaging of Target Scattering Centers via Wide-Angle Incidence in Radar Networks. <i>Sensors</i> 2025, <i>25</i>, 6802.

Sensors (Basel, Switzerland)·2026
Same journal

Enhancing Unsupervised Multi-Source Domain Adaptation for Person Re-Identification via Mixture of Experts and Graph-Based Relation.

Sensors (Basel, Switzerland)·2026
Same journal

Development of an Instrumented Glove for Palmar Pressure Assessment in Kayakers.

Sensors (Basel, Switzerland)·2026
Same journal

Development and Experimental Validation of an Autonomous IoT-Based Monitoring System for Real-Time Water Quality Assessment in the Amazon River.

Sensors (Basel, Switzerland)·2026
Same journal

Semi-Supervised Adversarial Learning Framework for Controller Area Network Bus Intrusion Detection.

Sensors (Basel, Switzerland)·2026
Same journal

Smart Optimization Method for Safety Signs in Innovative Manufacturing Environments Integrating Industrial Field IoT Sensors and Knowledge Graphs.

Sensors (Basel, Switzerland)·2026
See all related articles

This study introduces a novel source-free semi-supervised domain adaptation framework for object detection in CCTV systems. The method enhances detection accuracy by fusing pseudo-labels and using static adversarial regularization, improving performance under domain shifts.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Artificial Intelligence

Background:

  • Object detection in Closed-Circuit Television (CCTV) systems suffers from performance degradation due to domain gaps.
  • Privacy concerns and data regulations necessitate source-free learning approaches, limiting data reuse.
  • Existing methods struggle with domain adaptation and privacy constraints in real-world CCTV applications.

Purpose of the Study:

  • To develop a stable and effective source-free semi-supervised domain adaptation framework for CCTV object detection.
  • To address performance degradation caused by domain shifts and privacy limitations.
  • To improve the reliability and accuracy of object detection in challenging real-world scenarios.

Main Methods:

  • Proposed a Mean Teacher-based framework integrating pseudo-label fusion from weakly and strongly augmented views for robust pseudo-labels.
Keywords:
CCTVdomain adaptationobject detectionsemi-supervised learningsource-free

More Related Videos

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

9.5K

Related Experiment Videos

Last Updated: Jan 13, 2026

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

1.0K
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

9.5K
  • Implemented static adversarial regularization (SAR) with a frozen adversarial head for stable domain-invariance constraints.
  • Utilized a time-varying exponential weighting strategy to balance labeled and unlabeled target data during training.

    • Main Results:

    • Achieved an average improvement of 7.2% in mAP@0.5 over existing methods across four benchmark scenarios.
    • Demonstrated a 6.8% gain in a low-label setting using only 2% labeled target data.
    • Showcased an average improvement of 5.4% under challenging domain shifts, including clear-to-foggy adaptation and synthetic-to-real transfer.

    Conclusions:

    • The proposed source-free semi-supervised domain adaptation framework is effective and stable for CCTV object detection.
    • The method significantly enhances object detection performance under domain shifts while respecting privacy constraints.
    • This approach offers practical relevance for real-world CCTV systems facing domain variability and data privacy regulations.