One Dinomaly2 Detect Them All: A Unified Framework for Full-Spectrum Unsupervised Anomaly Detection

TL;DR

Dinomaly2 is a unified, minimalist framework for full-spectrum unsupervised anomaly detection that outperforms existing models across diverse data modalities, tasks, and domains with minimal modifications.

Contribution

It introduces Dinomaly2, the first truly unified framework for image anomaly detection that achieves superior performance through a simple, scalable, and adaptable design across multiple scenarios.

Findings

Achieves 99.9% I-AUROC on MVTec-AD

Surpasses previous full-shot models with only 8 normal examples per class

Demonstrates state-of-the-art results across 12 benchmarks and multiple modalities

Abstract

Unsupervised anomaly detection (UAD) has evolved from building specialized single-class models to unified multi-class models, yet existing multi-class models significantly underperform the most advanced one-for-one counterparts. Moreover, the field has fragmented into specialized methods tailored to specific scenarios (multi-class, 3D, few-shot, etc.), creating deployment barriers and highlighting the need for a unified solution. In this paper, we present Dinomaly2, the first unified framework for full-spectrum image UAD, which bridges the performance gap in multi-class models while seamlessly extending across diverse data modalities and task settings. Guided by the "less is more" philosophy, we demonstrate that the orchestration of five simple element achieves superior performance in a standard reconstruction-based framework. This methodological minimalism enables natural extension across diverse tasks without modification, establishing that simplicity is the foundation of true universality. Extensive experiments on 12 UAD benchmarks demonstrate Dinomaly2's full-spectrum superiority across multiple modalities (2D, multi-view, RGB-3D, RGB-IR), task settings (single-class, multi-class, inference-unified multi-class, few-shot) and application domains (industrial, biological, outdoor). For example, our multi-class model achieves unprecedented 99.9% and 99.3% image-level (I-) AUROC on MVTec-AD and VisA respectively. For multi-view and multi-modal inspection, Dinomaly2 demonstrates state-of-the-art performance with minimum adaptations. Moreover, using only 8 normal examples per class, our method surpasses previous full-shot models, achieving 98.7% and 97.4% I-AUROC on MVTec-AD and VisA. The combination of minimalistic design, computational scalability, and universal applicability positions Dinomaly2 as a unified solution for the full spectrum of real-world anomaly detection applications.