PatchEAD: Unifying Industrial Visual Prompting Frameworks for Patch-Exclusive Anomaly Detection
This work addresses the problem of fragmented visual prompting for industrial anomaly detection, offering a training-free solution for quick deployment, though it is incremental in unifying existing approaches.
The authors tackled the fragmentation of visual prompting in industrial anomaly detection by proposing PatchEAD, a unified patch-focused framework that achieves superior few-shot and zero-shot performance without textual features.
Industrial anomaly detection is increasingly relying on foundation models, aiming for strong out-of-distribution generalization and rapid adaptation in real-world deployments. Notably, past studies have primarily focused on textual prompt tuning, leaving the intrinsic visual counterpart fragmented into processing steps specific to each foundation model. We aim to address this limitation by proposing a unified patch-focused framework, Patch-Exclusive Anomaly Detection (PatchEAD), enabling training-free anomaly detection that is compatible with diverse foundation models. The framework constructs visual prompting techniques, including an alignment module and foreground masking. Our experiments show superior few-shot and batch zero-shot performance compared to prior work, despite the absence of textual features. Our study further examines how backbone structure and pretrained characteristics affect patch-similarity robustness, providing actionable guidance for selecting and configuring foundation models for real-world visual inspection. These results confirm that a well-unified patch-only framework can enable quick, calibration-light deployment without the need for carefully engineered textual prompts.