We present a comprehensive experimental study on pretrained feature extractors for visual out-of-distribution (OOD) detection. We examine several setups, based on the availability of labels or image captions and using different combinations of in- and out-distributions. Intriguingly, we find that (i) contrastive language-image pretrained models achieve state-of-the-art unsupervised out-of-distribution performance using nearest neighbors feature similarity as the OOD detection score, (ii) supervised state-of-the-art OOD detection performance can be obtained without in-distribution fine-tuning, (iii) even top-performing billion-scale vision transformers trained with natural language supervision fail at detecting adversarially manipulated OOD images. Finally, we argue whether new benchmarks for visual anomaly detection are needed based on our experiments. Using the largest publicly available vision transformer, we achieve state-of-the-art performance across all $18$ reported OOD benchmarks, including an AUROC of 87.6\% (9.2\% gain, unsupervised) and 97.4\% (1.2\% gain, supervised) for the challenging task of CIFAR100 $\rightarrow$ CIFAR10 OOD detection. The code will be open-sourced.

该文章从预训练功能提取器、视觉超出分布（OOD）检测、语言图像预训练模型、对抗性操纵的OOD图像和视觉异常检测等方面进行了全面实验研究，证明了使用最近邻特征相似度作为OOD检测得分的对比语言-图像预训练模型实现了最先进的无监督OOD性能，同时可以在无内部分布微调的情况下获得有监督的最先进OOD检测性能，并指出需要基于实验研究建立新的视觉异常检测基准。

对比语言-图像预训练（CLIP）模型是强大的超出分布检测器