Anomaly Detection on VisA

Metric: Segmentation AUROC (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide augmentations

Sort:

#	Model↕	Segmentation AUROC▼	Augmentations	Paper	Date↕	Code
1	Dinomaly ViT-L (model-unified multi-class)	99.1	No	Dinomaly: The Less Is More Philosophy in Multi-C...	2024-05-23	Code
2	HETMM	99.1	No	Hard-normal Example-aware Template Mutual Matchi...	2023-03-28	Code
3	URD	99.1	Yes	Unlocking the Potential of Reverse Distillation ...	2024-12-10	Code
4	INP-Former ViT-B (model-unified multi-class)	98.9	No	Exploring Intrinsic Normal Prototypes within a S...	2025-03-04	Code
5	DiffusionAD	98.9	No	DiffusionAD: Norm-guided One-step Denoising Diff...	2023-03-15	Code
6	UniNet	98.8	No	-	-	Code
7	GLASS	98.8	No	A Unified Anomaly Synthesis Strategy with Gradie...	2024-07-12	Code
8	RealNet	98.8	Yes	RealNet: A Feature Selection Network with Realis...	2024-03-09	Code
9	AnomalyDINO-S (full-shot)	98.8	No	AnomalyDINO: Boosting Patch-based Few-shot Anoma...	2024-05-23	Code
10	MuSc (zero-shot)	98.8	No	MuSc: Zero-Shot Industrial Anomaly Classificatio...	2024-01-30	Code
11	FAIRnoDTD	98.7	No	FAIR: Frequency-aware Image Restoration for Indu...	2023-09-13	Code
12	GLAD	98.6	No	GLAD: Towards Better Reconstruction with Global ...	2024-06-11	Code
13	PBAS	98.6	No	Progressive Boundary Guided Anomaly Synthesis fo...	2024-12-23	Code
14	CRAS	98.4	No	Center-aware Residual Anomaly Synthesis for Mult...	2025-05-23	Code
15	ReContrast	98.2	No	-	-	-
16	AnomalyDINO-S (4-shot)	98.2	No	AnomalyDINO: Boosting Patch-based Few-shot Anoma...	2024-05-23	Code
17	AnomalyDINO-S (2-shot)	98	No	AnomalyDINO: Boosting Patch-based Few-shot Anoma...	2024-05-23	Code
18	D3AD	97.9	No	Dynamic Addition of Noise in a Diffusion Model f...	2024-01-09	Code
19	AnomalyDINO-S (1-shot)	97.8	No	AnomalyDINO: Boosting Patch-based Few-shot Anoma...	2024-05-23	Code
20	DDAD	97.6	No	Anomaly Detection with Conditioned Denoising Dif...	2023-05-25	Code
21	VCP-CLIP	95.7	Yes	VCP-CLIP: A visual context prompting model for z...	2024-07-17	Code
22	AdaCLIP	95.5	Yes	AdaCLIP: Adapting CLIP with Hybrid Learnable Pro...	2024-07-22	Code
23	AnomalyCLIP	95.5	Yes	AnomalyCLIP: Object-agnostic Prompt Learning for...	2023-10-29	Code
24	APRIL-GAN	94.2	Yes	APRIL-GAN: A Zero-/Few-Shot Anomaly Classificati...	2023-05-27	Code

#1Dinomaly ViT-L (model-unified multi-class)
99.1
Segmentation AUROC· 2024-05-23
Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection Code
#2HETMMSOTA
99.1
Segmentation AUROC· 2023-03-28
Hard-normal Example-aware Template Mutual Matching for Industrial Anomaly Detection Code
#3URD
99.1
Segmentation AUROC· Augmentations· 2024-12-10
Unlocking the Potential of Reverse Distillation for Anomaly Detection Code
#4INP-Former ViT-B (model-unified multi-class)
98.9
Segmentation AUROC· 2025-03-04
Exploring Intrinsic Normal Prototypes within a Single Image for Universal Anomaly Detection Code
#5DiffusionADSOTA
98.9
Segmentation AUROC· 2023-03-15
DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection Code
#6UniNet
98.8
Segmentation AUROC
No paperCode
#7GLASS
98.8
Segmentation AUROC· 2024-07-12
A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localization Code
#8RealNet
98.8
Segmentation AUROC· Augmentations· 2024-03-09
RealNet: A Feature Selection Network with Realistic Synthetic Anomaly for Anomaly Detection Code
#9AnomalyDINO-S (full-shot)
98.8
Segmentation AUROC· 2024-05-23
AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2 Code
#10MuSc (zero-shot)
98.8
Segmentation AUROC· 2024-01-30
MuSc: Zero-Shot Industrial Anomaly Classification and Segmentation with Mutual Scoring of the Unlabeled Images Code
#11FAIRnoDTD
98.7
Segmentation AUROC· 2023-09-13
FAIR: Frequency-aware Image Restoration for Industrial Visual Anomaly Detection Code
#12GLAD
98.6
Segmentation AUROC· 2024-06-11
GLAD: Towards Better Reconstruction with Global and Local Adaptive Diffusion Models for Unsupervised Anomaly Detection Code
#13PBAS
98.6
Segmentation AUROC· 2024-12-23
Progressive Boundary Guided Anomaly Synthesis for Industrial Anomaly Detection Code
#14CRAS
98.4
Segmentation AUROC· 2025-05-23
Center-aware Residual Anomaly Synthesis for Multi-class Industrial Anomaly Detection Code
#15ReContrast
98.2
Segmentation AUROC
No paper
#16AnomalyDINO-S (4-shot)
98.2
Segmentation AUROC· 2024-05-23
AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2 Code
#17AnomalyDINO-S (2-shot)
98
Segmentation AUROC· 2024-05-23
AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2 Code
#18D3AD
97.9
Segmentation AUROC· 2024-01-09
Dynamic Addition of Noise in a Diffusion Model for Anomaly Detection Code
#19AnomalyDINO-S (1-shot)
97.8
Segmentation AUROC· 2024-05-23
AnomalyDINO: Boosting Patch-based Few-shot Anomaly Detection with DINOv2 Code
#20DDAD
97.6
Segmentation AUROC· 2023-05-25
Anomaly Detection with Conditioned Denoising Diffusion Models Code
#21VCP-CLIP
95.7
Segmentation AUROC· Augmentations· 2024-07-17
VCP-CLIP: A visual context prompting model for zero-shot anomaly segmentation Code
#22AdaCLIP
95.5
Segmentation AUROC· Augmentations· 2024-07-22
AdaCLIP: Adapting CLIP with Hybrid Learnable Prompts for Zero-Shot Anomaly Detection Code
#23AnomalyCLIP
95.5
Segmentation AUROC· Augmentations· 2023-10-29
AnomalyCLIP: Object-agnostic Prompt Learning for Zero-shot Anomaly Detection Code
#24APRIL-GAN
94.2
Segmentation AUROC· Augmentations· 2023-05-27
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 Code