EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Zero-Shot Industrial Anomaly Detection via CLIP-DINOv2 Multimodal Fusion and Stabilized Attention Pooling

Junjie Jiang, Zongxiang He, Anping Wan, Khalil AL-Bukhaiti, Kaiyang Wang, Peiyi Zhu, Xiaomin Cheng

Electronics · 05.12.2025 · 14 (24) · 4785

DOI: 10.3390/electronics14244785 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 05.12.2025. Verfasst von Junjie Jiang und 6 weiteren Autor:innen. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 42 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Joint Fund of Zhejiang BaiMa Lake Laboratory, Scientific Research Foundation of Hangzhou City University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Industrial visual inspection demands high-precision anomaly detection amid scarce annotations and unseen defects. This paper introduces a zero-shot framework leveraging multimodal feature fusion and stabilized attention pooling. CLIP’s global semantic embeddings are hierarchically aligned with DINOv2’s multi-scale structural features via a Dual-Modality Attention (DMA) mechanism, enabling effective cross-modal knowledge transfer for capturing macro- and micro-anomalies. A Stabilized Attention-based Pooling (SAP) module adaptively aggregates discriminative representations using self-generated anomaly heatmaps, enhancing localization accuracy and mitigating feature dilution. Trained solely in auxiliary datasets with multi-task segmentation and contrastive losses, the approach requires no target-domain samples. Extensive evaluation across seven benchmarks (MVTec AD, VisA, BTAD, MPDD, KSDD, DAGM, DTD-Synthetic) demonstrates state-of-the-art performance, achieving 93.4% image-level AUROC, 94.3% AP, 96.9% pixel-level AUROC, and 92.4% AUPRO on average. Ablation studies confirm the efficacy of DMA and SAP, while qualitative results highlight superior boundary precision and noise suppression. The framework offers a scalable, annotation-efficient solution for real-world industrial anomaly detection.

Förderung

  • National Natural Science Foundation of China
  • Joint Fund of Zhejiang BaiMa Lake Laboratory
  • Scientific Research Foundation of Hangzhou City University
  • Key R&D Program of Zhejiang Province
  • Leading Goose R&D Program of Zhejiang Province
  • Ningbo Science and Technology Innovation 2035 Key Technology Breakthrough Program

Zitieren

APA

Jiang, J., He, Z., Wan, A., AL-Bukhaiti, K., Wang, K., Zhu, P., & Cheng, X. (2025). Zero-Shot Industrial Anomaly Detection via CLIP-DINOv2 Multimodal Fusion and Stabilized Attention Pooling. Electronics, 14(24), 4785. https://doi.org/10.3390/electronics14244785

Vancouver

Jiang J, He Z, Wan A, AL-Bukhaiti K, Wang K, Zhu P, et al. Zero-Shot Industrial Anomaly Detection via CLIP-DINOv2 Multimodal Fusion and Stabilized Attention Pooling. Electronics. 2025;14(24):4785. doi:10.3390/electronics14244785

BibTeX
@article{Jiang2025,
  author = {Jiang, Junjie and He, Zongxiang and Wan, Anping and AL-Bukhaiti, Khalil and Wang, Kaiyang and Zhu, Peiyi and Cheng, Xiaomin},
  title = {Zero-Shot Industrial Anomaly Detection via CLIP-DINOv2 Multimodal Fusion and Stabilized Attention Pooling},
  journal = {Electronics},
  year = {2025},
  volume = {14},
  number = {24},
  pages = {4785},
  doi = {10.3390/electronics14244785}
}
Kostenlos starten →