用新方法让视觉模型直接零样本检测异常，效果远超现有技术。
AnomalyVFM -- Transforming Vision Foundation Models into Zero-Shot Anomaly Detectors
- 构建合成数据集并高效适配视觉模型
- 在9个数据集上图像级准确率达94.1%，领先3.3个百分点
- 适合想用预训练模型做异常检测的研究者
零样本异常检测旨在不依赖任何领域内训练图像的情况下，识别并定位图像中的异常区域。尽管近期方法利用视觉语言模型（如CLIP）迁移高层概念知识，但基于纯视觉基础模型（如DINOv2）的方法性能仍落后。我们认为这一差距源于两个实际问题：(i) 现有辅助异常检测数据集多样性不足；(ii) 视觉模型适配策略过于浅显。为此，我们提出AnomalyVFM，一个通用且高效的框架，可将任意预训练视觉基础模型转化为强大的零样本异常检测器。该方法结合稳健的三阶段合成数据生成方案与参数高效适配机制，采用低秩特征适配器和置信度加权像素损失。这些组件共同使现代视觉基础模型显著超越当前最优方法。具体而言，以RADIO为骨干网络时，AnomalyVFM在9个多样化数据集上实现平均图像级AUROC为94.1%，比之前方法提升3.3个百分点。
原文摘要 · Abstract (English)
Zero-shot anomaly detection aims to detect and localise abnormal regions in the image without access to any in-domain training images. While recent approaches leverage vision-language models (VLMs), such as CLIP, to transfer high-level concept knowledge, methods based on purely vision foundation models (VFMs), like DINOv2, have lagged behind in performance. We argue that this gap stems from two practical issues: (i) limited diversity in existing auxiliary anomaly detection datasets and (ii) overly shallow VFM adaptation strategies. To address both challenges, we propose AnomalyVFM, a general and effective framework that turns any pretrained VFM into a strong zero-shot anomaly detector. Our approach combines a robust three-stage synthetic dataset generation scheme with a parameter-efficient adaptation mechanism, utilising low-rank feature adapters and a confidence-weighted pixel loss. Together, these components enable modern VFMs to substantially outperform current state-of-the-art methods. More specifically, with RADIO as a backbone, AnomalyVFM achieves an average image-level AUROC of 94.1% across 9 diverse datasets, surpassing previous methods by significant 3.3 percentage points. Project Page: https://maticfuc.github.io/anomaly_vfm/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。