Unsupervised Anomaly Detection

The objective of Unsupervised Anomaly Detection is to detect previously unseen rare objects or events without any prior knowledge about these. The only information available is that the percentage of anomalies in the dataset is small, usually less than 1%. Since anomalies are rare and unknown to the user at training time, anomaly detection in most cases boils down to the problem of modelling the normal data distribution and defining a measurement in this space in order to classify samples as anomalous or normal. In high-dimensional data such as images, distances in the original space quickly lose descriptive power (curse of dimensionality) and a mapping to some more suitable space is required.

Source: Unsupervised Learning of Anomaly Detection from Contaminated Image Data using Simultaneous Encoder Training

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 1–25 of 506 papers

Title	Date	Tasks	Status	Hype	Score
MambaAD: Exploring State Space Models for Multi-class Unsupervised Anomaly Detection	Apr 9, 2024	Anomaly DetectionDecoder	CodeCode Available	3	5
Dinomaly: The Less Is More Philosophy in Multi-Class Unsupervised Anomaly Detection	May 23, 2024	Anomaly DetectionMulti-class Anomaly Detection	CodeCode Available	3	5
Graph-Augmented Normalizing Flows for Anomaly Detection of Multiple Time Series	Feb 16, 2022	Anomaly DetectionDensity Estimation	CodeCode Available	3	5
Real-IAD: A Real-World Multi-View Dataset for Benchmarking Versatile Industrial Anomaly Detection	Mar 19, 2024	Anomaly DetectionBenchmarking	CodeCode Available	3	5
A Unified Anomaly Synthesis Strategy with Gradient Ascent for Industrial Anomaly Detection and Localization	Jul 12, 2024	Anomaly DetectionDefect Detection	CodeCode Available	3	5
3CAD: A Large-Scale Real-World 3C Product Dataset for Unsupervised Anomaly	Feb 9, 2025	Anomaly DetectionUnsupervised Anomaly Detection	CodeCode Available	2	5
TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data	Jan 18, 2022	Anomaly DetectionMeta-Learning	CodeCode Available	2	5
Unsupervised Continual Anomaly Detection with Contrastively-learned Prompt	Jan 2, 2024	Anomaly DetectionAnomaly Segmentation	CodeCode Available	2	5
SoftPatch+: Fully Unsupervised Anomaly Classification and Segmentation	Dec 30, 2024	Anomaly ClassificationAnomaly Detection	CodeCode Available	2	5
Anomaly Detection via Reverse Distillation from One-Class Embedding	Jan 26, 2022	Anomaly Classification	CodeCode Available	2	5
SoftPatch: Unsupervised Anomaly Detection with Noisy Data	Mar 21, 2024	Anomaly DetectionUnsupervised Anomaly Detection	CodeCode Available	2	5
Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications	Feb 12, 2018	Anomaly DetectionUnsupervised Anomaly Detection	CodeCode Available	2	5
Correcting Deviations from Normality: A Reformulated Diffusion Model for Multi-Class Unsupervised Anomaly Detection	Mar 25, 2025	Anomaly DetectionUnsupervised Anomaly Detection	CodeCode Available	2	5
Towards Total Recall in Industrial Anomaly Detection	Jun 15, 2021	3D Anomaly DetectionAnomaly Classification	CodeCode Available	2	5
A Unified Model for Multi-class Anomaly Detection	Jun 8, 2022	Anomaly DetectionAnomaly Localization	CodeCode Available	2	5
AnoDDPM: Anomaly Detection With Denoising Diffusion Probabilistic Models Using Simplex Noise	Jun 30, 2022	Anomaly DetectionDenoising	CodeCode Available	2	5
CostFilter-AD: Enhancing Anomaly Detection through Matching Cost Filtering	May 2, 2025	Anomaly DetectionUnsupervised Anomaly Detection	CodeCode Available	2	5
DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection	Mar 15, 2023	Anomaly DetectionDenoising	CodeCode Available	2	5
EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies	Mar 25, 2023	Anomaly DetectionComputational Efficiency	CodeCode Available	2	5
GLAD: Towards Better Reconstruction with Global and Local Adaptive Diffusion Models for Unsupervised Anomaly Detection	Jun 11, 2024	Anomaly DetectionDenoising	CodeCode Available	2	5
TadGAN: Time Series Anomaly Detection Using Generative Adversarial Networks	Sep 16, 2020	Anomaly DetectionBenchmarking	CodeCode Available	2	5
Continual Learning for Unsupervised Anomaly Detection in Continuous Auditing of Financial Accounting Data	Dec 25, 2021	Anomaly Detectioncontinual anomaly detection	CodeCode Available	1	5
Constrained Contrastive Distribution Learning for Unsupervised Anomaly Detection and Localisation in Medical Images	Mar 5, 2021	Anomaly DetectionContrastive Learning	CodeCode Available	1	5
A SAM-guided Two-stream Lightweight Model for Anomaly Detection	Feb 29, 2024	Anomaly DetectionDecoder	CodeCode Available	1	5
AnoViT: Unsupervised Anomaly Detection and Localization with Vision Transformer-based Encoder-Decoder	Mar 21, 2022	Anomaly DetectionAnomaly Localization	CodeCode Available	1	5

Show:10 25 50

← PrevPage 1 of 21Next →

All datasets AnoShift SMAP Vehicle Claims Fashion-MNIST KolektorSDD2 20NEWS AeBAD-S Caltech-101 DAGM2007 ECG5000 KolektorSDD MNIST

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	ACR-NTL (zero-shot, test anomaly ratio=1%)	ROC-AUC FAR	62.5	—	Unverified
2	ACR-DSVDD (zero-shot, anomaly ratio=1%)	ROC-AUC FAR	62	—	Unverified
3	ACR-NTL (zero-shot, test anomaly ratio=20%)	ROC-AUC FAR	62	—	Unverified
4	ACR-DSVDD (zero-shot, anomaly ratio=20%)	ROC-AUC FAR	59.1	—	Unverified
5	COPOD	ROC-AUC FAR	50.42	—	Unverified
6	OC-SVM	ROC-AUC FAR	49.57	—	Unverified
7	SO-GAAL	ROC-AUC FAR	49.35	—	Unverified
8	ECOD Li et al. (2022)	ROC-AUC FAR	49.19	—	Unverified
9	LOF	ROC-AUC FAR	34.96	—	Unverified
10	deepSVDD	ROC-AUC FAR	34.53	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DFM (flow matching)	F1	94.1	—	Unverified
2	ContextFlow++ (Glow-based)	F1	93.62	—	Unverified
3	TranAd	F1	89.15	—	Unverified
4	MTAD-GAT	F1	88.8	—	Unverified
5	CAE-M	F1	88.27	—	Unverified
6	OmniAnomaly	F1	87.28	—	Unverified
7	Glow	F1	86.05	—	Unverified
8	GDN	F1	85.18	—	Unverified
9	USAD	F1	81.86	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SOM	AUC	65.43	—	Unverified
2	Isolation Forest	AUC	59.42	—	Unverified
3	Latent Outlier Exposure	AUC	58.59	—	Unverified
4	NeuTraL-AD	AUC	57.03	—	Unverified
5	RSRAE	AUC	55.38	—	Unverified
6	SOM-DAGMM	AUC	53.82	—	Unverified
7	Local Outlier Factor	AUC	52.86	—	Unverified
8	One Class Support Vector Machines	AUC	51.68	—	Unverified
9	DAGMM	AUC	51.22	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	RSRAE	AUC-ROC	0.85	—	Unverified
2	RSRAE	AUC (outlier ratio = 0.5)	0.83	—	Unverified
3	RSRAE	AUC-ROC	0.75	—	Unverified
4	RSRAE	AUC-ROC	0.69	—	Unverified
5	RSRAE	AUC-ROC	0.69	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Semi-orthogonal	Segmentation AUROC	98.1	—	Unverified
2	WeakREST-Un	Segmentation AP	76.9	—	Unverified
3	DSR	Segmentation AP	61.4	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	RSRAE	AUC (outlier ratio = 0.5)	0.83	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MSFR	Detection AUROC	87.1	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	RSRAE	AUC (outlier ratio = 0.5)	0.77	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DiffusionAD	Detection AUROC	99.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	VRAE+SVM	AUC	0.98	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Semi-orthogonal	Segmentation AUROC	96	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	LVAD	AUROC	0.94	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DyEdgeGAT	AUC	0.8	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	RSRAE	AUC (outlier ratio = 0.5)	0.85	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	TranAD	Precision	92.62	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	LVAD	AUC-ROC	1	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DyEdgeGAT	AUC	0.83	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	P-CAE W-MSE (Tilted View)	AUROC	78.1	—	Unverified