VarAD: Lightweight High-Resolution Image Anomaly Detection via Visual Autoregressive Modeling

TL;DR

VarAD introduces a lightweight, high-resolution image anomaly detection method using visual autoregressive modeling, effectively capturing global information and outperforming existing methods on multiple datasets.

Contribution

The paper proposes VarAD, a novel autoregressive approach for high-resolution image anomaly detection that leverages multi-hierarchy token sequences and a lightweight model for superior performance.

Findings

Achieves state-of-the-art results on four datasets.

Maintains lightweight architecture suitable for high-resolution images.

Demonstrates effectiveness on real-world inspection data.

Abstract

This paper addresses a practical task: High-Resolution Image Anomaly Detection (HRIAD). In comparison to conventional image anomaly detection for low-resolution images, HRIAD imposes a heavier computational burden and necessitates superior global information capture capacity. To tackle HRIAD, this paper translates image anomaly detection into visual token prediction and proposes VarAD based on visual autoregressive modeling for token prediction. Specifically, VarAD first extracts multi-hierarchy and multi-directional visual token sequences, and then employs an advanced model, Mamba, for visual autoregressive modeling and token prediction. During the prediction process, VarAD effectively exploits information from all preceding tokens to predict the target token. Finally, the discrepancies between predicted tokens and original tokens are utilized to score anomalies. Comprehensive experiments on four publicly available datasets and a real-world button inspection dataset demonstrate that the proposed VarAD achieves superior high-resolution image anomaly detection performance while maintaining lightweight, rendering VarAD a viable solution for HRIAD. Code is available at \href{https://github.com/caoyunkang/VarAD}{\url{https://github.com/caoyunkang/VarAD}}.