One-for-More: Continual Diffusion Model for Anomaly Detection

TL;DR

This paper introduces a continual diffusion model for anomaly detection that addresses faithfulness hallucination and catastrophic forgetting, using gradient projection and SVD-based memory efficiency, achieving state-of-the-art results.

Contribution

It proposes a novel continual diffusion framework with gradient projection and SVD-based memory reduction for improved anomaly detection.

Findings

Achieved first place in 17 out of 18 settings on MVTec and VisA datasets.

Effectively mitigated hallucination and forgetting in diffusion models.

Enhanced anomaly detection with an anomaly-masked network.

Abstract

With the rise of generative models, there is a growing interest in unifying all tasks within a generative framework. Anomaly detection methods also fall into this scope and utilize diffusion models to generate or reconstruct normal samples when given arbitrary anomaly images. However, our study found that the diffusion model suffers from severe ``faithfulness hallucination'' and ``catastrophic forgetting'', which can't meet the unpredictable pattern increments. To mitigate the above problems, we propose a continual diffusion model that uses gradient projection to achieve stable continual learning. Gradient projection deploys a regularization on the model updating by modifying the gradient towards the direction protecting the learned knowledge. But as a double-edged sword, it also requires huge memory costs brought by the Markov process. Hence, we propose an iterative singular value decomposition method based on the transitive property of linear representation, which consumes tiny memory and incurs almost no performance loss. Finally, considering the risk of ``over-fitting'' to normal images of the diffusion model, we propose an anomaly-masked network to enhance the condition mechanism of the diffusion model. For continual anomaly detection, ours achieves first place in 17/18 settings on MVTec and VisA. Code is available at https://github.com/FuNz-0/One-for-More