| UniNet: A Contrastive Learning-guided Unified Framework with Feature Selection for Anomaly Detection | ✓ Link | 99.8 | 93.9 | | 93.9 | 98.8 | UniNet | 2025-02-28 |
| GLAD: Towards Better Reconstruction with Global and Local Adaptive Diffusion Models for Unsupervised Anomaly Detection | ✓ Link | 99.5 | 94.3 | 98.3 | 94.3 | 98.6 | GLAD | 2024-06-11 |
| UniNet: A Contrastive Learning-guided Unified Framework with Feature Selection for Anomaly Detection | ✓ Link | 99.15 | | 98.29 | | | UniNet(model-unified multi-class) | 2025-02-28 |
| Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection | ✓ Link | 98.9 | 94.8 | 96.1 | 94.8 | 99.1 | Dinomaly ViT-L (model-unified multi-class) | 2024-05-23 |
| Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detection | ✓ Link | 98.9 | 94.4 | 96.6 | 94.4 | 98.9 | INP-Former ViT-B (model-unified multi-class) | 2025-03-04 |
| Anomaly Detection with Conditioned Denoising Diffusion Models | ✓ Link | 98.9 | 92.7 | | | 97.6 | DDAD | 2023-05-25 |
| DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection | ✓ Link | 98.8 | 96.0 | | 96.0 | 98.9 | DiffusionAD | 2023-03-15 |
| A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localization | ✓ Link | 98.8 | 92.8 | | | 98.8 | GLASS | 2024-07-12 |
| TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detection | ✓ Link | 98.7 | 94.7 | | | | TransFusion | 2023-11-16 |
| EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies | ✓ Link | 98.1 | 94.0 | | | | EfficientAD-M | 2023-03-25 |
| Hard-normal Example-aware Template Mutual Matching for Industrial Anomaly Detection | ✓ Link | 98.1 | | | | 99.1 | HETMM | 2023-03-28 |
| RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection | ✓ Link | 97.8 | | | | 98.8 | RealNet | 2024-03-09 |
| Progressive Boundary Guided Anomaly Synthesis for Industrial Anomaly Detection | ✓ Link | 97.7 | | | 93.3 | 98.6 | PBAS | 2024-12-23 |
| AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2 | ✓ Link | 97.6 | 96.1 | | | 98.8 | AnomalyDINO-S (full-shot) | 2024-05-23 |
| EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies | ✓ Link | 97.5 | 93.1 | | | | EfficientAD-S | 2023-03-25 |
| []() | | 97.5 | | | 92.6 | 98.2 | ReContrast | |
| FAIR: Frequency-aware Image Restoration for Industrial Visual Anomaly Detection | ✓ Link | 97.1 | 91.2 | | | 98.7 | FAIRnoDTD | 2023-09-13 |
| Center-aware Residual Anomaly Synthesis for Multi-class Industrial Anomaly Detection | ✓ Link | 97.0 | | | | 98.4 | CRAS | 2025-05-23 |
| Unlocking the Potential of Reverse Distillation for Anomaly Detection | ✓ Link | 96.5 | 95.1 | | | 99.1 | URD | 2024-12-10 |
| Dynamic Addition of Noise in a Diffusion Model for Anomaly Detection | ✓ Link | 96.0 | 94.1 | | | 97.9 | D3AD | 2024-01-09 |
| Asymmetric Student-Teacher Networks for Industrial Anomaly Detection | ✓ Link | 94.9 | 81.5 | | | | AST | 2022-10-14 |
| Reconstruction from edge image combined with color and gradient difference for industrial surface anomaly detection | ✓ Link | 94.2 | 90.7 | | | | EdgRec | 2022-10-26 |
| SuperSimpleNet: Unifying Unsupervised and Supervised Learning for Fast and Reliable Surface Defect Detection | ✓ Link | 93.4 | 87.4 | | 87.4 | | SuperSimpleNet | 2024-08-06 |
| Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddings | ✓ Link | 93.2 | | | | | Student-Teacher | 2019-11-06 |
| MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled Images | ✓ Link | 92.8 | 92.7 | | 92.7 | 98.8 | MuSc (zero-shot) | 2024-01-30 |
| AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2 | ✓ Link | 92.6 | 94.1 | | | 98.2 | AnomalyDINO-S (4-shot) | 2024-05-23 |
| CFA: Coupled-hypersphere-based Feature Adaptation for Target-Oriented Anomaly Localization | ✓ Link | 92.0 | 55.1 | | | | CFA | 2022-06-09 |
| CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flows | ✓ Link | 91.5 | | | | | CFLOW | 2021-07-27 |
| AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2 | ✓ Link | 89.7 | 93.4 | | | 98 | AnomalyDINO-S (2-shot) | 2024-05-23 |
| Beyond Dents and Scratches: Logical Constraints in Unsupervised Anomaly Detection and Localization | | 89.1 | 83.7 | | | | GCAD | 2022-02-22 |
| SPot-the-Difference Self-Supervised Pre-training for Anomaly Detection and Segmentation | ✓ Link | 87.8 | | | | | SPD | 2022-07-28 |
| AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2 | ✓ Link | 87.4 | 92.5 | | | 97.8 | AnomalyDINO-S (1-shot) | 2024-05-23 |
| WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation | ✓ Link | 87.3 | 87.6 | | 87.6 | | WinCLIP+ (4-shot) | 2023-03-26 |
| AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detection | ✓ Link | 85.8 | | | | 95.5 | AdaCLIP | 2024-07-22 |
| WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation | ✓ Link | 84.6 | 86.2 | | 86.2 | | WinCLIP+ (2-shot) | 2023-03-26 |
| WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation | ✓ Link | 83.8 | 85.1 | | 85.1 | | WinCLIP+ (1-shot) | 2023-03-26 |
| Student-Teacher Feature Pyramid Matching for Anomaly Detection | ✓ Link | 83.3 | 62.0 | | | | STPM | 2021-03-07 |
| Sub-Image Anomaly Detection with Deep Pyramid Correspondences | ✓ Link | 82.1 | 65.9 | | | | SPADE | 2020-05-05 |
| AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detection | ✓ Link | 82.1 | | | 87.0 | 95.5 | AnomalyCLIP | 2023-10-29 |
| AnoDDPM: Anomaly Detection With Denoising Diffusion Probabilistic Models Using Simplex Noise | ✓ Link | 78.2 | 60.5 | | | | AnoDDPM | 2022-06-30 |
| WinCLIP: Zero-/Few-Shot Anomaly Classification and Segmentation | ✓ Link | 78.1 | 56.8 | | 56.8 | | WinCLIP (0-shot) | 2023-03-26 |
| APRIL-GAN: A Zero-/Few-Shot Anomaly Classification and Segmentation Method for CVPR 2023 VAND Workshop Challenge Tracks 1&2: 1st Place on Zero-shot AD and 4th Place on Few-shot AD | ✓ Link | 78.0 | | 32.3 | 86.8 | 94.2 | APRIL-GAN | 2023-05-27 |
| PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization | ✓ Link | | 85.9 | | | | PaDiM | 2020-11-17 |
| DRAEM -- A discriminatively trained reconstruction embedding for surface anomaly detection | ✓ Link | | 73.1 | | | | DRAEM | 2021-08-17 |
| Anomaly Detection via Reverse Distillation from One-Class Embedding | ✓ Link | | 70.9 | | | | Reverse Distillation | 2022-01-26 |
| DSR -- A dual subspace re-projection network for surface anomaly detection | ✓ Link | | 68.1 | | | | DSR | 2022-08-02 |
| Anomaly localization by modeling perceptual features | ✓ Link | | 67.9 | | | | FAVAE | 2020-08-12 |
| FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows | ✓ Link | | 59.8 | | | | FastFlow | 2021-11-15 |
| Segment Any Anomaly without Training via Hybrid Prompt Regularization | ✓ Link | | | 27.07 | | | SAA+ | 2023-05-18 |
| VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentation | ✓ Link | | | | 90.7 | 95.7 | VCP-CLIP | 2024-07-17 |