UniSpector: Towards Universal Open-set Defect Recognition via Spectral-Contrastive Visual Prompting

arXiv cs.CV / 4/6/2026

📰 NewsSignals & Early TrendsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper argues that industrial inspection systems need open-set defect recognition to detect unprecedented anomalies, since most current methods assume a closed set of defects.
It proposes UniSpector, a visual prompting framework designed to avoid prompt embedding collapse by building a semantically structured and transferable prompt topology.
UniSpector introduces a Spatial-Spectral Prompt Encoder for orientation-invariant, fine-grained representations and a Contrastive Prompt Encoder that regularizes prompts onto a semantically organized angular manifold.
The method uses Prompt-guided Query Selection to generate adaptive object queries aligned with the prompt, aiming to improve defect localization performance.
The authors present Inspect Anything, the first benchmark for visual-prompt-based open-set defect localization, where UniSpector achieves at least 19.7% (AP50b) and 15.8% (AP50m) gains over baselines and supports a retraining-free inspection paradigm for evolving environments.

Abstract

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.