COFT-AD: COntrastive Fine-Tuning for Few-Shot Anomaly Detection
- 12 March 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 33 (10577149) , 2090-2103
Abstract
Existing approaches towards anomaly detection (AD) often rely on a substantial amount of anomaly-free data to train representation and density models. However, large anomaly-free datasets may not always be available before the inference stage; in which case an anomaly detection model must be trained with only a handful of normal samples, a.k.a. few-shot anomaly detection (FSAD). In this paper, we propose a novel methodology to address the challenge of FSAD which incorporates two important techniques. Firstly, we employ a model pre-trained on a large source dataset to initialize model weights. Secondly, to ameliorate the covariate shift between source and target domains, we adopt contrastive training to fine-tune on the few-shot target domain data. To learn suitable representations for the downstream AD task, we additionally incorporate cross-instance positive pairs to encourage a tight cluster of the normal samples, and negative pairs for better separation between normal and synthesized negative samples. We evaluate few-shot anomaly detection on on 3 controlled AD tasks and 4 real-world AD tasks to demonstrate the effectiveness of the proposed method.Keywords
All Related Versions
Funding Information
- Agency for Science, Technology and Research (A*STAR) through the Advanced Manufacturing and Engineering (AME) Programmatic Funds (A20H6b0151)
This publication has 34 references indexed in Scilit:
- Anomaly Detection via Few-Shot Learning on Normality2023
- Registration Based Few-Shot Anomaly Detection2022
- Patch SVDD: Patch-Level SVDD for Anomaly Detection and Segmentation2021
- Do Better ImageNet Models Transfer Better?2019
- OCGAN: One-Class Novelty Detection Using GANs With Constrained Latent Representations2019
- Deep visual domain adaptation: A surveyNeurocomputing, 2018
- Surface defect saliency of magnetic tileThe Visual Computer, 2018
- Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery2017
- Deep Residual Learning for Image Recognition2016
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015