Kernel-Aware Graph Prompt Learning for Few-Shot Anomaly Detection

TL;DR

This paper introduces KAG-prompt, a kernel-aware graph prompt learning framework that captures cross-layer visual feature relationships to improve few-shot anomaly detection accuracy.

Contribution

It proposes a novel hierarchical graph model that reasons about cross-layer visual features and introduces a multi-level information fusion method for better anomaly prediction.

Findings

Achieves state-of-the-art results on MVTecAD and VisA datasets.

Effectively captures cross-layer contextual information for anomaly detection.

Improves both image-level and pixel-level anomaly detection performance.

Abstract

Few-shot anomaly detection (FSAD) aims to detect unseen anomaly regions with the guidance of very few normal support images from the same class. Existing FSAD methods usually find anomalies by directly designing complex text prompts to align them with visual features under the prevailing large vision-language model paradigm. However, these methods, almost always, neglect intrinsic contextual information in visual features, e.g., the interaction relationships between different vision layers, which is an important clue for detecting anomalies comprehensively. To this end, we propose a kernel-aware graph prompt learning framework, termed as KAG-prompt, by reasoning the cross-layer relations among visual features for FSAD. Specifically, a kernel-aware hierarchical graph is built by taking the different layer features focusing on anomalous regions of different sizes as nodes, meanwhile, the relationships between arbitrary pairs of nodes stand for the edges of the graph. By message passing over this graph, KAG-prompt can capture cross-layer contextual information, thus leading to more accurate anomaly prediction. Moreover, to integrate the information of multiple important anomaly signals in the prediction map, we propose a novel image-level scoring method based on multi-level information fusion. Extensive experiments on MVTecAD and VisA datasets show that KAG-prompt achieves state-of-the-art FSAD results for image-level/pixel-level anomaly detection. Code is available at https://github.com/CVL-hub/KAG-prompt.git.