MAEDAY: MAE for few and zero shot AnomalY-Detection

2022/11/25 by Schwartz, Eli, Arbelle, Assaf, Karlinsky, Leonid +4 · 4 citations
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2211.14307

Abstract

We propose using Masked Auto-Encoder (MAE), a transformer model self-supervisedly trained on image inpainting, for anomaly detection (AD). Assuming anomalous regions are harder to reconstruct compared with normal regions. MAEDAY is the first image-reconstruction-based anomaly detection method that utilizes a pre-trained model, enabling its use for Few-Shot Anomaly Detection (FSAD). We also show the same method works surprisingly well for the novel tasks of Zero-Shot AD (ZSAD) and Zero-Shot Foreign Object Detection (ZSFOD), where no normal samples are available. Code is available at https://github.com/EliSchwartz/MAEDAY .

Cited by

• AgentIAD: Agentic Industrial Anomaly Detection via Adaptive Memory Augmentation
• Quantized FCA: Efficient Zero-Shot Texture Anomaly Detection
• DictAS: A Framework for Class-Generalizable Few-Shot Anomaly Segmentation via Dictionary Lookup
• A Comprehensive Survey for Real-World Industrial Defect Detection: Challenges, Approaches, and Prospects

Related

• Foundation Visual Encoders Are Secretly Few-Shot Anomaly Detectors
• AnoRefiner: Anomaly-Aware Group-Wise Refinement for Zero-Shot Industrial Anomaly Detection
• 3D Masked Autoencoders with Application to Anomaly Detection in Non-Contrast Enhanced Breast MRI
• Registration based Few-Shot Anomaly Detection
• AFR-CLIP: Enhancing Zero-Shot Industrial Anomaly Detection with Stateless-to-Stateful Anomaly Feature Rectification