ShortcutBreaker: Low-Rank Noisy Bottleneck and Frequency Filtering Block for Multi-Class Unsupervised Anomaly Detection

TL;DR

ShortcutBreaker introduces a low-rank noisy bottleneck and global perturbation attention to enhance multi-class unsupervised anomaly detection, effectively preventing identity shortcuts and improving detection accuracy across diverse datasets.

Contribution

It presents a novel unified framework with theoretical and practical innovations to address identity shortcuts in MUAD, outperforming previous methods.

Findings

Achieves up to 99.8% image-level AUROC on benchmark datasets.

Effectively prevents trivial identity reproduction in features.

Outperforms previous MUAD methods across multiple scenarios.

Abstract

Multi-class unsupervised anomaly detection (MUAD) has garnered growing research interest, as it seeks to develop a unified model for anomaly detection across multiple classes, i.e., eliminating the need to train separate models for distinct objects and thereby saving substantial computational resources. Under the MUAD setting, while advanced Transformer-based architectures have brought significant performance improvements, identity shortcuts persist: they directly copy inputs to outputs, narrowing the gap in reconstruction errors between normal and abnormal cases, and thereby making the two harder to distinguish. Therefore, we propose ShortcutBreaker, a novel unified feature-reconstruction framework for MUAD tasks, featuring two key innovations to address the issue of shortcuts. First, drawing on matrix rank inequality, we design a low-rank noisy bottleneck (LRNB) to project highdimensional features into a low-rank latent space, and theoretically demonstrate its capacity to prevent trivial identity reproduction. Second, leveraging ViTs global modeling capability instead of merely focusing on local features, we incorporate a global perturbation attention to prevent information shortcuts in the decoders. Extensive experiments are performed on four widely used anomaly detection benchmarks, including three industrial datasets (MVTec-AD, ViSA, and Real-IAD) and one medical dataset (Universal Medical). The proposed method achieves a remarkable image-level AUROC of 99.8%, 98.9%, 90.6%, and 87.8% on these four datasets, respectively, consistently outperforming previous MUAD methods across different scenarios Our code will be released..