EUVIMEDEUVIMED entdecken
Crossref

Unsupervised Continual Anomaly Detection with Contrastively-Learned Prompt

Jiaqi Liu, Kai Wu, Qiang Nie, Ying Chen, Bin-Bin Gao, Yong Liu, Jinbao Wang, Chengjie Wang, Feng Zheng

Proceedings of the AAAI Conference on Artificial Intelligence · 24.03.2024 · 38 (4) · 3639-3647

DOI: 10.1609/aaai.v38i4.28153 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 24.03.2024. Verfasst von Jiaqi Liu und 8 weiteren Autor:innen. Die Arbeit wurde bisher 44-mal zitiert (Crossref).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unsupervised Anomaly Detection (UAD) with incremental training is crucial in industrial manufacturing, as unpredictable defects make obtaining sufficient labeled data infeasible. However, continual learning methods primarily rely on supervised annotations, while the application in UAD is limited due to the absence of supervision. Current UAD methods train separate models for different classes sequentially, leading to catastrophic forgetting and a heavy computational burden. To address this issue, we introduce a novel Unsupervised Continual Anomaly Detection framework called UCAD, which equips the UAD with continual learning capability through contrastively-learned prompts. In the proposed UCAD, we design a Continual Prompting Module (CPM) by utilizing a concise key-prompt-knowledge memory bank to guide task-invariant 'anomaly' model predictions using task-specific 'normal' knowledge. Moreover, Structure-based Contrastive Learning (SCL) is designed with the Segment Anything Model (SAM) to improve prompt learning and anomaly segmentation results. Specifically, by treating SAM's masks as structure, we draw features within the same mask closer and push others apart for general feature representations. We conduct comprehensive experiments and set the benchmark on unsupervised continual anomaly detection and segmentation, demonstrating that our method is significantly better than anomaly detection methods, even with rehearsal training. The code will be available at https://github.com/shirowalker/UCAD.

Zitieren

APA

Liu, J., Wu, K., Nie, Q., Chen, Y., Gao, B., Liu, Y., Wang, J., Wang, C., & Zheng, F. (2024). Unsupervised Continual Anomaly Detection with Contrastively-Learned Prompt. Proceedings of the AAAI Conference on Artificial Intelligence, 38(4), 3639-3647. https://doi.org/10.1609/aaai.v38i4.28153

Vancouver

Liu J, Wu K, Nie Q, Chen Y, Gao B, Liu Y, et al. Unsupervised Continual Anomaly Detection with Contrastively-Learned Prompt. Proceedings of the AAAI Conference on Artificial Intelligence. 2024;38(4):3639-3647. doi:10.1609/aaai.v38i4.28153

BibTeX
@article{Liu2024,
  author = {Liu, Jiaqi and Wu, Kai and Nie, Qiang and Chen, Ying and Gao, Bin-Bin and Liu, Yong and Wang, Jinbao and Wang, Chengjie and Zheng, Feng},
  title = {Unsupervised Continual Anomaly Detection with Contrastively-Learned Prompt},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2024},
  volume = {38},
  number = {4},
  pages = {3639--3647},
  doi = {10.1609/aaai.v38i4.28153}
}
Kostenlos starten →