Abstract:Unsupervised anomaly detection (UAD) attracts a lot of research interest and drives widespread applications, where only anomaly-free samples are available for training. Some UAD applications intend to further locate the anomalous regions without any anomaly information. Although the absence of anomalous samples and annotations deteriorates the UAD performance, an inconspicuous yet powerful statistics model, the normalizing flows, is appropriate for anomaly detection and localization in an unsupervised fashion. The flow-based probabilistic models, only trained on anomaly-free data, can efficiently distinguish unpredictable anomalies by assigning them much lower likelihoods than normal data. Nevertheless, the size variation of unpredictable anomalies introduces another inconvenience to the flow-based methods for high-precision anomaly detection and localization. To generalize the anomaly size variation, we propose a novel Multi-Scale Flow-based framework dubbed MSFlow composed of asymmetrical parallel flows followed by a fusion flow to exchange multi-scale perceptions. Moreover, different multi-scale aggregation strategies are adopted for image-wise anomaly detection and pixel-wise anomaly localization according to the discrepancy between them. The proposed MSFlow is evaluated on three anomaly detection datasets, significantly outperforming existing methods. Notably, on the challenging MVTec AD benchmark, our MSFlow achieves a new state-of-the-art with a detection AUORC score of up to 99.7%, localization AUCROC score of 98.8%, and PRO score of 97.1%. The reproducible code is available at <a class="link-external link-https" href="https://github.com/cool-xuan/msflow" rel="external noopener nofollow">this https URL</a>.

PyramidFlow: High-Resolution Defect Contrastive Localization using Pyramid Normalizing Flow

A Multi-scale Normalizing Flow Network for Surface Defects Detection

Fully Convolutional Cross-Scale-Flows for Image-based Defect Detection

Visual Defect Inspection of Metal Part Surface via Deformable Convolution and Concatenate Feature Pyramid Neural Networks

FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows

PET-3DFlow: A Normalizing Flow Based Method for 3D PET Anomaly Detection

CTFlow: Mitigating Effects of Computed Tomography Acquisition and Reconstruction with Normalizing Flows

PU-Flow: a Point Cloud Upsampling Network with Normalizing Flows

Learning Regularized Multi-Scale Feature Flow for High Dynamic Range Imaging

Computer Vision and Normalizing Flow-Based Defect Detection

NFCF: Industrial Surface Anomaly Detection with Normalizing Flow Cross-Fitting Network

FlowCLAS: Enhancing Normalizing Flow Via Contrastive Learning For Anomaly Segmentation

VQ-Flow: Taming Normalizing Flows for Multi-Class Anomaly Detection via Hierarchical Vector Quantization

CT-3DFlow : Leveraging 3D Normalizing Flows for Unsupervised Detection of Pathological Pulmonary CT scans

Unsupervised Anomaly Detection and Localization Based on Two-Hierarchy Normalizing Flow

MSFlow: Multiscale Flow-Based Framework for Unsupervised Anomaly Detection

FRAnomaly: flow-based rapid anomaly detection from images

Cross-Attention Regression Flow for Defect Detection

CL-Flow:Strengthening the Normalizing Flows by Contrastive Learning for Better Anomaly Detection

MSFlow: Multi-Scale Flow-based Framework for Unsupervised Anomaly Detection

Visual Anomaly Detection via Dual-Attention Transformer and Discriminative Flow