Robust Distribution Alignment for Industrial Anomaly Detection under Distribution Shift
This addresses robustness in quality control for industrial applications, though it appears incremental as it builds on existing memory-bank-based methods.
The paper tackles the problem of industrial anomaly detection under distribution shifts like lighting variations, proposing a method that optimizes a robust Sinkhorn distance on limited target data to enhance generalization. It demonstrates superior results compared to state-of-the-art methods on 2D and 3D benchmarks with simulated shifts.
Anomaly detection plays a crucial role in quality control for industrial applications. However, ensuring robustness under unseen domain shifts such as lighting variations or sensor drift remains a significant challenge. Existing methods attempt to address domain shifts by training generalizable models but often rely on prior knowledge of target distributions and can hardly generalise to backbones designed for other data modalities. To overcome these limitations, we build upon memory-bank-based anomaly detection methods, optimizing a robust Sinkhorn distance on limited target training data to enhance generalization to unseen target domains. We evaluate the effectiveness on both 2D and 3D anomaly detection benchmarks with simulated distribution shifts. Our proposed method demonstrates superior results compared with state-of-the-art anomaly detection and domain adaptation methods.