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

Discrepancy-Guided Complementary Fusion for Unsupervised Multimodal Anomaly Detection.

Taehui Lee1, Seyoung Jeong1, Sang Jun Lee1

  • 1Division of Electronic Engineering, Jeonbuk National University, 567 Baekje-daero, Deokjin-gu, Jeonju 54896, Republic of Korea.

Sensors (Basel, Switzerland)
|June 26, 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

Spatial-Semantic Object Relation Graph Networks for Vehicle Attachment Detection in Automatic Car Wash System.

Sensors (Basel, Switzerland)·2026
Same author

ADBM: Adversarial Diffusion Bridge Model for Denoising of 3D Point Cloud Data.

Sensors (Basel, Switzerland)·2025
Same author

RGB-Based Visual-Inertial Odometry via Knowledge Distillation from Self-Supervised Depth Estimation with Foundation Models.

Sensors (Basel, Switzerland)·2025
Same author

MECO: Mixture-of-Expert Codebooks for Multiple Dense Prediction Tasks.

Sensors (Basel, Switzerland)·2025
Same author

Neural Radiance Fields for Fisheye Driving Scenes Using Edge-Aware Integrated Depth Supervision.

Sensors (Basel, Switzerland)·2024
Same author

Uncertainty-Aware Depth Network for Visual Inertial Odometry of Mobile Robots.

Sensors (Basel, Switzerland)·2024
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 new multimodal anomaly detection framework for industrial inspection. It improves defect detection by fusing complementary information from different sensors, outperforming existing methods.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Industrial Automation

Background:

  • Industrial inspection relies on detecting subtle defects, often challenging with single sensors.
  • Multimodal inspection enhances object representation by combining appearance and geometric data.
  • Early fusion in multimodal methods can dilute anomaly signals and degrade performance.

Purpose of the Study:

  • To develop an improved unsupervised multimodal anomaly detection framework.
  • To address the limitations of early fusion strategies in existing methods.
  • To enhance the detection and localization of subtle defects in industrial settings.

Main Methods:

  • Proposed a reconstruction-based unsupervised multimodal anomaly detection framework.
  • Integrated Discrepancy-Guided Complementary Fusion (DGCF) to exploit cross-modal discrepancies.
Keywords:
feature extractionfeature-level fusionindustrial anomaly detectionmulti-sensor fusionmultimodal anomaly detectionunsupervised learning

Related Experiment Videos

  • Utilized Noise to Feature (N2F) for feature reconstruction regularization and learning robust normal representations.

    • Main Results:

    • Achieved high performance on MVTec 3D-AD (97.3% I-AUROC, 99.6% P-AUROC, 97.6% AUPRO).
    • Demonstrated strong results on Eyecandies dataset (94.8% I-AUROC, 98.6% P-AUROC, 93.4% AUPRO).
    • The proposed method effectively reduces anomaly smoothing and improves detection accuracy.

    Conclusions:

    • The novel framework significantly enhances multimodal anomaly detection capabilities.
    • DGCF and N2F integration effectively address challenges of feature fusion and representation learning.
    • The method shows great promise for reliable industrial inspection and defect identification.