Hypergraph-Enhanced Training-Free and Language-Free Few-Shot Anomaly Detection

2026-05-11

AI Summary

This work addresses key limitations in few-shot anomaly detection—such as reliance on task-specific training, language supervision, handcrafted prompts, and poor cross-domain robustness—by proposing a purely visual, training-free, and prompt-free approach. Built upon DINOv3 features, the method introduces a sparse hyper-matching mechanism and a dual-branch image scoring strategy, enabling robust cross-domain inference through support-aware CLS matching, hypergraph reasoning, and spatial-global feature fusion. Evaluated on six industrial and medical datasets, the proposed method achieves state-of-the-art performance, significantly outperforming existing few-shot anomaly detection techniques.

Abstract

Few-shot anomaly detection (FSAD) has made significant strides, yet existing methods still face critical challenges: (i) dependence on task- or dataset-specific training/fine-tuning, (ii) reliance on language supervision or carefully hand-crafted prompts, and (iii) limited robustness across domains. In this paper, we introduce HyperFSAD, a novel FSAD framework that is training-free, language-free, and robust across domains, offering a powerful solution to these challenges. Built upon DINOv3 and a hypergraph-based inference mechanism, our approach performs inference without any task-specific optimization or text prompts, while remaining competitive. Specifically, we replace sensitive nearest-neighbor / top-$n$ matching with \textbf{Sparse Hyper Matching}: \textit{sparsemax} first selects the most relevant support patches, which are then aggregated into a \textit{hyperedge} as compact normal evidence to suppress background noise and distractors. We further introduce \textbf{Dual-Branch Image Scoring}, which fuses \emph{spatial anomaly evidence} from the patch-grid anomaly map with \emph{global semantic deviation} captured by support-aware CLS matching, yielding a robust image-level anomaly score in a strictly visual manner. Notably, all components of HyperFSAD are purely visual, eliminating the need for labor-intensive hand-crafted text prompts. Under the stringent training-free and language-free setting, HyperFSAD achieves state-of-the-art performance across six datasets spanning four industrial datasets (MVTecAD, VisA, MPDD, BTAD) and two medical datasets (RESC, BraTS).

Problem

• few-shot anomaly detection
• training-free
• language-free
• domain robustness
• anomaly detection

Innovation

• Hypergraph
• Training-Free
• Language-Free
• Few-Shot Anomaly Detection
• Sparse Hyper Matching