awesome-anomaly-synthesis

🔥🔥🔥 Contributions to our repository are welcome. Feel free to categorize the papers.

Taxonomy of Industrial Image Anomaly Synthesis

Hand-Crafted Model

Self-Contained

• Cutpaste: Self-supervised learning for anomaly detection and localization [CVPR 2021][unofficial code]
• Self-supervised learning for anomaly detection with dynamic local augmentation [IEEE Access 2021]
• A Study on Data Augmentation Techniques for Visual Defect Detection in Manufacturing [BUAV 2021]
• Anomaly Detection of GAN Industrial Image Based on Attention Feature Fusion [Sensors 2022]
• Self-Supervised Learning for Industrial Image Anomaly Detection by Simulating Anomalous Samples [IJCIS 2023]
• REB: Reducing Biases in Representation for Industrial Anomaly Detection [KBS 2023][code]
• Natural Synthetic Anomalies for Self-supervised Anomaly Detection and Localization (NSA) [ECCV 2022][code]

External-Dependent

• Draem-a discriminatively trained reconstruction embedding for surface anomaly detection [ICCV 2021][code]
• MemSeg: A semi-supervised method for image surface defect detection using differences and commonalities [EAAI 2023][unofficial code]
• DeSTSeg: Segmentation Guided Denoising Student-Teacher for Anomaly Detection [CVPR 2023][code]

Inpainting-Based

• Superpixel masking and inpainting for self-supervised anomaly detection (SMAI) [BMVC 2020]
• Reconstruction by inpainting for visual anomaly detection (RIAD) [PR 2021]
• Iterative image inpainting with structural similarity mask for anomaly detection (I3AD) [2021]
• Inpainting transformer for anomaly detection (InTra) [ICIAP 2022]
• Cutout as augmentation in contrastive learning for detecting burn marks in plastic granules [JSSS 2024]
• AMI-Net: Adaptive Mask Inpainting Network for Industrial Anomaly Detection and Localization [TASE 2024]

Distribution Hypothesis-Based Models

Prior-Dependent

• Manifolds for unsupervised visual anomaly detection [2020]
• Anomaly Detection using Score-based Perturbation Resilience [ICCV 2023]
• Progressive Boundary Guided Anomaly Synthesis for Industrial Anomaly Detection [TCSVT 2024][code]
• A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localization (GLASS) [ECCV 2024][code]

Data-Driven

• DSR: A dual subspace re-projection network for surface anomaly detection [ECCV 2022][code]
• A Unified Model for Multi-class Anomaly Detection (UniAD) [NeurIPS 2022] [code]
• SimpleNet: A Simple Network for Image Anomaly Detection and Localization [CVPR 2023][code]
• GLAD: Towards Better Reconstruction with Global and Local Adaptive Diffusion Models for Unsupervised Anomaly Detection [ECCV 2024][code]
• SuperSimpleNet: Unifying Unsupervised and Supervised Learning for Fast and Reliable Surface Defect Detection [ICPR 2024][code]

GM-Based Models

Full-Image Synthesis

• Multistage GAN for fabric defect detection [TIP 2019]
• A new contrastive GAN with data augmentation for surface defect recognition under limited data (Con-GAN) [TIM 2022]
• Few-shot defect image generation via defect-aware feature manipulation (DFMGAN) [AAAI 2023][code]
• FEGAN: A Feature Extraction Based Approach For GAN Anomaly Detection And Localization (FEGAN) [IEEE Access 2024]
• Defect Spectrum: A Granular Look of Large-Scale Defect Datasets with Rich Semantics [ECCV 2024][data]

Full-Image Translation

• Defect image sample generation with GAN for improving defect recognition [TASE 2020]
• Defect-GAN: High-fidelity defect synthesis for automated defect inspection [WACV 2021]
• A new cycle-consistent adversarial networks with attention mechanism for surface defect classification with small samples (AttenCGAN) [TII 2022]

Local Anomaly Synthesis

• Mask-guided generation method for industrial defect images with non-uniform structures [Machines 2022]
• Diversified and Multi-Class Controllable Industrial Defect Synthesis for Data Augmentation and Transfer [CVPRW 2023]
• RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection [CVPR 2024][code]

Vison Language Models-Based

One-Stage Generation

• CAGEN: Controllable Anomaly Generator using Diffusion Model [ICASSP 2024]
• CUT: A Controllable, Universal, and Training-Free Visual Anomaly Generation Framework [2024]
• AnomalyXFusion: Multi-modal Anomaly Synthesis with Diffusion [2024][code]
• AnomalyControl: Learning Cross-modal Semantic Features for Controllable Anomaly Synthesis [2024]
• SeaS: Few-shot Industrial Anomaly Image Generation with Separation and Sharing Fine-tuning [2024]

Multi-Stage Generation

• AnomalyDiffusion: Few-Shot Anomaly Image Generation with Diffusion Model [AAAI 2024][code]
• Few-shot defect image generation based on consistency modeling [ECCV 2024]
• DualAnoDiff: Dual-Interrelated Diffusion Model for Few-Shot Anomaly Image Generation [2024][code]
• A Novel Approach to Industrial Defect Generation through Blended Latent Diffusion Model with Online Adaptation [2024][code]