Abstract:Diffusion models have shown superior performance on unsupervised anomaly detection tasks. Since trained with normal data only, diffusion models tend to reconstruct normal counterparts of test images with certain noises added. However, these methods treat all potential anomalies equally, which may cause two main problems. From the global perspective, the difficulty of reconstructing images with different anomalies is uneven. Therefore, instead of utilizing the same setting for all samples, we propose to predict a particular denoising step for each sample by evaluating the difference between image contents and the priors extracted from diffusion models. From the local perspective, reconstructing abnormal regions differs from normal areas even in the same image. Theoretically, the diffusion model predicts a noise for each step, typically following a standard Gaussian distribution. However, due to the difference between the anomaly and its potential normal counterpart, the predicted noise in abnormal regions will inevitably deviate from the standard Gaussian distribution. To this end, we propose introducing synthetic abnormal samples in training to encourage the diffusion models to break through the limitation of standard Gaussian distribution, and a spatial-adaptive feature fusion scheme is utilized during inference. With the above modifications, we propose a global and local adaptive diffusion model (abbreviated to GLAD) for unsupervised anomaly detection, which introduces appealing flexibility and achieves anomaly-free reconstruction while retaining as much normal information as possible. Extensive experiments are conducted on three commonly used anomaly detection datasets (MVTec-AD, MPDD, and VisA) and a printed circuit board dataset (PCB-Bank) we integrated, showing the effectiveness of the proposed method.

LGFDR: local and global feature denoising reconstruction for unsupervised anomaly detection

Dual-Path Reconstruction Guided Segmentation Network for Unsupervised Anomaly Detection and Localization.

A Diffusion-Based Framework for Multi-Class Anomaly Detection

AEGLR-Net: Attention Enhanced Global-Local Refined Network for Accurate Detection of Car Body Surface Defects

Anomaly Detection via Progressive Reconstruction and Hierarchical Feature Fusion

DFR: Deep Feature Reconstruction for Unsupervised Anomaly Segmentation

Unsupervised Anomaly Detection via Normal Feature-Enhanced Reverse Teacher–Student Distillation

Focus Your Distribution: Coarse-to-Fine Non-Contrastive Learning for Anomaly Detection and Localization

Latent Feature Reconstruction for Unsupervised Anomaly Detection.

Pull & Push: Leveraging Differential Knowledge Distillation for Efficient Unsupervised Anomaly Detection and Localization

Noise-to-Norm Reconstruction for Industrial Anomaly Detection and Localization

Context-aware Feature Reconstruction for Class-Incremental Anomaly Detection and Localization

Multi-feature Reconstruction Network using Crossed-mask Restoration for Unsupervised Industrial Anomaly Detection

GLAD: Towards Better Reconstruction with Global and Local Adaptive Diffusion Models for Unsupervised Anomaly Detection

Cascade RDN: Towards Accurate Localization in Industrial Visual Anomaly Detection with Structural Anomaly Generation

Attention-Guided Perturbation for Unsupervised Image Anomaly Detection

ALMRR: Anomaly Localization Mamba on Industrial Textured Surface with Feature Reconstruction and Refinement

Unsupervised Anomaly Detection for Surface Defects with Dual-Siamese Network

Unified Model Based on Reinforced Feature Reconstruction for Metro Track Anomaly Detection

Foreign Object Debris Detection for Airport Pavement Images Based on Image Random Masking and Semantic Feature Matching

Revisiting Deep Feature Reconstruction for Logical and Structural Industrial Anomaly Detection