Segment Any Anomaly without Training via Hybrid Prompt Regularization

TL;DR

This paper introduces SAA+, a zero-shot anomaly segmentation framework that leverages foundation models and hybrid prompt regularization to achieve state-of-the-art results without domain-specific training.

Contribution

The work proposes a novel zero-shot anomaly segmentation method combining foundation models with hybrid prompts for improved generalization.

Findings

Achieves state-of-the-art zero-shot performance on multiple benchmarks.

Effectively leverages foundation models for anomaly localization.

Demonstrates robustness across diverse anomaly patterns.

Abstract

We present a novel framework, i.e., Segment Any Anomaly + (SAA+), for zero-shot anomaly segmentation with hybrid prompt regularization to improve the adaptability of modern foundation models. Existing anomaly segmentation models typically rely on domain-specific fine-tuning, limiting their generalization across countless anomaly patterns. In this work, inspired by the great zero-shot generalization ability of foundation models like Segment Anything, we first explore their assembly to leverage diverse multi-modal prior knowledge for anomaly localization. For non-parameter foundation model adaptation to anomaly segmentation, we further introduce hybrid prompts derived from domain expert knowledge and target image context as regularization. Our proposed SAA+ model achieves state-of-the-art performance on several anomaly segmentation benchmarks, including VisA, MVTec-AD, MTD, and KSDD2, in the zero-shot setting. We will release the code at \href{https://github.com/caoyunkang/Segment-Any-Anomaly}{https://github.com/caoyunkang/Segment-Any-Anomaly}.