Kota Dohi

Robustness of Zero-Shot CLIP to Contextual Variations in Industrial Anomaly Detection: A Factorized Sensitivity Analysis

Automated optical inspection is crucial for maintaining quality in modern manufacturing. While deep-learning-based anomaly detection (AD) has shown promise, its deployment is often hindered by “contextual variations,” namely benign environmental changes such as lighting or camera noise that are irrelevant to product de...