UniSpector：面向通用开放集缺陷识别的光谱对比视觉提示
UniSpector: Towards Universal Open-set Defect Recognition via Spectral-Contrastive Visual Prompting
提出基于视觉提示的开放集缺陷定位框架，利用空间-光谱提示编码器提取方向不变细粒度特征，并通过对比提示编码器将提示空间正则化为语义组织的角度流形，为开放词表检测和少样本缺陷识别提供了可复用的提示拓扑设计思路。
原文依据 we propose UniSpector, which shifts the focus from naive prompt-to-region matching to the principled design of a semantically structured and transferable prompt topology.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论尽管工业检测系统应能够识别前所未有的缺陷，但大多数现有方法在闭集假设下运行，这阻碍了它们检测新异常。虽然视觉提示为工业检测提供了一种可扩展的替代方案，但现有方法常因类内方差大和类间差异细微而遭受提示嵌入崩溃。为解决此问题，我们提出UniSpector，它将焦点从朴素的提示到区域匹配转向对语义结构化且可迁移的提示拓扑的原则性设计。UniSpector采用空间-光谱提示编码器提取方向不变的细粒度表示；这些表示为对比提示编码器提供了坚实基础，以显式地将提示空间正则化为语义组织的角度流形。此外，提示引导的查询选择生成与提示对齐的自适应对象查询。我们引入了Inspect Anything，这是首个基于视觉提示的开放集缺陷定位基准，其中UniSpector在AP50b和AP50m上分别以至少19.7%和15.8%显著优于基线。这些结果表明，我们的方法为持续演进的工业环境提供了一种可扩展、免重新训练的检测范式，同时为通用视觉提示的设计提供了关键见解。
英文原文摘要 1423 字符
Although industrial inspection systems should be capable of recognizing unprecedented defects, most existing approaches operate under a closed-set assumption, which prevents them from detecting novel anomalies. While visual prompting offers a scalable alternative for industrial inspection, existing methods often suffer from prompt embedding collapse due to high intra-class variance and subtle inter-class differences. To resolve this, we propose UniSpector, which shifts the focus from naive prompt-to-region matching to the principled design of a semantically structured and transferable prompt topology. UniSpector employs the Spatial-Spectral Prompt Encoder to extract orientation-invariant, fine-grained representations; these serve as a solid basis for the Contrastive Prompt Encoder to explicitly regularize the prompt space into a semantically organized angular manifold. Additionally, Prompt-guided Query Selection generates adaptive object queries aligned with the prompt. We introduce Inspect Anything, the first benchmark for visual-prompt-based open-set defect localization, where UniSpector significantly outperforms baselines by at least 19.7% and 15.8% in AP50b and AP50m, respectively. These results show that our method enable a scalable, retraining-free inspection paradigm for continuously evolving industrial environments, while offering critical insights into the design of generic visual prompting.