DiffusionAD: Norm-Guided One-Step Denoising Diffusion for Anomaly Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2025-08 · 47 (8) · 7140-7152
DOI: 10.1109/tpami.2025.3570494 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2025-08. Verfasst von Hui Zhang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 97-mal zitiert (Crossref) und nennt 62 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Anomaly detection has garnered extensive applications in real industrial manufacturing due to its remarkable effectiveness and efficiency. However, previous generative-based models have been limited by suboptimal reconstruction quality, hampering their overall performance. We introduce DiffusionAD, a novel anomaly detection pipeline comprising a reconstruction sub-network and a segmentation sub-network. A fundamental enhancement lies in our reformulation of the reconstruction process using a diffusion model into a noise-to-norm paradigm. Here, the anomalous region loses its distinctive features after being disturbed by Gaussian noise and is subsequently reconstructed into an anomaly-free one. Afterward, the segmentation sub-network predicts pixel-level anomaly scores based on the similarities and discrepancies between the input image and its anomaly-free reconstruction. Additionally, given the substantial decrease in inference speed due to the iterative denoising nature of diffusion models, we revisit the denoising process and introduce a rapid one-step denoising paradigm. This paradigm achieves hundreds of times acceleration while preserving comparable reconstruction quality. Furthermore, considering the diversity in the manifestation of anomalies, we propose a norm-guided paradigm to integrate the benefits of multiple noise scales, enhancing the fidelity of reconstructions. Comprehensive evaluations on four standard and challenging benchmarks reveal that DiffusionAD outperforms current state-of-the-art approaches and achieves comparable inference speed, demonstrating the effectiveness and broad applicability of the proposed pipeline.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Zhang, H., Wang, Z., Zeng, D., Wu, Z., & Jiang, Y. G. (2025). DiffusionAD: Norm-Guided One-Step Denoising Diffusion for Anomaly Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence, 47(8), 7140-7152. https://doi.org/10.1109/tpami.2025.3570494
Vancouver
Zhang H, Wang Z, Zeng D, Wu Z, Jiang YG. DiffusionAD: Norm-Guided One-Step Denoising Diffusion for Anomaly Detection. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2025;47(8):7140-7152. doi:10.1109/tpami.2025.3570494
BibTeX
@article{Zhang2025,
author = {Zhang, Hui and Wang, Zheng and Zeng, Dan and Wu, Zuxuan and Jiang, Yu-Gang},
title = {DiffusionAD: Norm-Guided One-Step Denoising Diffusion for Anomaly Detection},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2025},
volume = {47},
number = {8},
pages = {7140--7152},
doi = {10.1109/tpami.2025.3570494}
}
EUVIMED