Enhancing Zero-Shot Anomaly Detection: CLIP-SAM Collaboration with Cascaded Prompts
Hou, Xu, Li et al.
Recently, the powerful generalization ability exhibited by foundation models has brought forth new solutions for zero-shot anomaly segmentation tasks. However, guiding these foundation models correctly to address downstream tasks remains a challenge. This paper proposes a novel two-stage framework, for zero-shot anomaly segmentation tasks in industrial anomaly detection. This framework excellently leverages the powerful anomaly localization capability of CLIP and the boundary perception ability of SAM.(1) To mitigate SAM's inclination towards object segmentation, we propose the Co-Feature Point Prompt Generation (PPG) module. This module collaboratively utilizes CLIP and SAM to generate positive and negative point prompts, guiding SAM to focus on segmenting anomalous regions rather than the entire object. (2) To further optimize SAM's segmentation results and mitigate rough boundaries and isolated noise, we introduce the Cascaded Prompts for SAM (CPS) module. This module employs hybrid prompts cascaded with a lightweight decoder of SAM, achieving precise segmentation of anomalous regions. Across multiple datasets, consistent experimental validation demonstrates that our approach achieves state-of-the-art zero-shot anomaly segmentation results. Particularly noteworthy is our performance on the Visa dataset, where we outperform the state-of-the-art methods by 10.3\% and 7.7\% in terms of {$F_1$-max} and AP metrics, respectively.
本論文は、産業異常検出におけるゼロショット異常セグメンテーション課題に対する新規な二段階フレームワークを提案している。本フレームワークはCLIPの強力な異常定位能力とSAMの境界認識能力を十分に活用している。Co-Feature Point Prompt Generation(PPG)モジュールとCascaded Prompts for SAM(CPS)モジュールを通じて、複数のデータセットで最先端のゼロショット異常セグメンテーション結果を実現し、特にVisAデータセットではF1-maxおよびAP指標が既存の最良手法と比べてそれぞれ10.3%および7.7%向上している。