Anomaly Detection

1007 papers with code • 59 benchmarks • 78 datasets

Anomaly Detection is a binary classification identifying unusual or unexpected patterns in a dataset, which deviate significantly from the majority of the data. The goal of anomaly detection is to identify such anomalies, which could represent errors, fraud, or other types of unusual events, and flag them for further investigation.

[Image source]: GAN-based Anomaly Detection in Imbalance Problems

Benchmarks

Add a Result

These leaderboards are used to track progress in Anomaly Detection

Dataset	Best Model	Compare
MVTec AD	EfficientAD (with early stopping)	See all
One-class CIFAR-10	CLIP (OE)	See all
MVTec LOCO AD	EfficientAD-M	See all
VisA	EfficientAD-M	See all
CUHK Avenue	AMSRC	See all
ShanghaiTech	AI-VAD	See all
Fishyscapes L&F	NFlowJS-GF (with extra inlier set: Vistas and Wilddash2)	See all
One-class CIFAR-100	Transformaly	See all
Unlabeled CIFAR-10 vs CIFAR-100	PsudoLabels ViT	See all
Fashion-MNIST	GAN based Anomaly Detection in Imbalance Problems	See all
One-class ImageNet-30	BCE-Clip (OE)	See all
UBnormal	STG-NF - Supervised	See all
Numenta Anomaly Benchmark	our	See all
Anomaly Detection on Unlabeled CIFAR-10 vs LSUN (Fix)	PsudoLabels ViT	See all
BTAD	CPR	See all
Fishyscapes	RPL+CoroCL	See all
Road Anomaly	RbA	See all
AeBAD-S	MMR	See all
AeBAD-V	MMR	See all
MNIST	GAN-based Anomaly Detection in Imbalance Problems	See all
UCSD Ped2	DMAD	See all
Hyper-Kvasir Dataset	CCD	See all
Leave-One-Class-Out CIFAR-10	BCE-CLIP	See all
Leave-One-Class-Out ImageNet-30	BCE-CLIP (OE)	See all
Anomaly Detection on Unlabeled ImageNet-30 vs CUB-200	PsudoLabels CLIP ViT	See all
Anomaly Detection on Anomaly Detection on Unlabeled ImageNet-30 vs Flowers-102	PsudoLabels CLIP ViT	See all
LAG	CCD	See all
Cats-and-Dogs	PANDA	See all
DIOR	PANDA-OE	See all
Lost and Found	Mask2Anomaly	See all
Surface Defect Saliency of Magnetic Tile	CS-Flow (unsupervised)	See all
UCSD Peds2	Background-Agnostic Framework	See all
ODDS	kNN	See all
UEA time-series datasets	SINBAD	See all
MVTEC AD textures	DualModel	See all
MVTec AD Textures Domain Generalization	FABLE	See all
Thyroid	DevNet	See all
MVTEC 3D-AD	CDO	See all
Corridor	Two-stream	See all
Vehicle Claims	Random Forest	See all
Census	DevNet	See all
MNIST-test	OGNET	See all
CIFAR-10	RCALAD	See all
AG News	DATE	See all
ASSIRA Cat Vs Dog	Shell-based Anomaly (supervisered)	See all
STL-10	Shell-based Anomaly (supervised)	See all
Forest CoverType	DIF	See all
Kaggle-Credit Card Fraud Dataset	DIF	See all
NB15-Analysis	DIF	See all
NB15-Backdoor	DIF	See all
NB15-DoS	DIF	See all
BottleCap	DFC	See all
kdd 99	PCA via oversampling	See all
MVTec 3D-AD (RGB)	CPR	See all
KDD Cup 1999	RCALAD	See all
MIT-BIH Arrhythmia Database	RCALAD	See all
Musk v1	RCALAD	See all
SVHN	RCALAD	See all
KSDD2	SAA+	See all

Show all 59 benchmarks

Collapse benchmarks

Libraries

Use these libraries to find Anomaly Detection models and implementations

xuhongzuo/DeepOD

12 papers

126

openvinotoolkit/anomalib

8 papers

2,376

yzhao062/pyod

4 papers

7,411

pygod-team/pygod

4 papers

1,040

Datasets

Subtasks

Anomaly Detection In Surveillance Videos

Image Manipulation Detection

Graph Anomaly Detection

Abnormal Event Detection In Video

Weakly-supervised Anomaly Detection

Self-Supervised Anomaly Detection

3D Anomaly Detection and Segmentation

RGB+3D Anomaly Detection and Segmentation

Contextual Anomaly Detection

Group Anomaly Detection

3D Anomaly Detection

Depth Anomaly Detection and Segmentation

RGB+Depth Anomaly Detection and Segmentation

Unsupervised Anomaly Detection In Sound

3D Anomaly Segmentation

Depth Anomaly Segmentation

3D + RGB Anomaly Segmentation

Depth + RGB Anomaly Segmentation

Depth + RGB Anomaly Detection

3D + RGB Anomaly Detection

DepthAnomaly Detection

Most implemented papers

Most implemented Social Latest No code

Auto-Encoding Variational Bayes

microsoft/recommenders • • 20 Dec 2013

First, we show that a reparameterization of the variational lower bound yields a lower bound estimator that can be straightforwardly optimized using standard stochastic gradient methods.

133

Paper
Code

EfficientAD: Accurate Visual Anomaly Detection at Millisecond-Level Latencies

openvinotoolkit/anomalib • • 25 Mar 2023

We train a student network to predict the extracted features of normal, i. e., anomaly-free training images.

Paper
Code

Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery

LeeDoYup/AnoGAN • • 17 Mar 2017

Obtaining models that capture imaging markers relevant for disease progression and treatment monitoring is challenging.

Paper
Code

PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization

openvinotoolkit/anomalib • • 17 Nov 2020

We present a new framework for Patch Distribution Modeling, PaDiM, to concurrently detect and localize anomalies in images in a one-class learning setting.

Paper
Code

Towards Total Recall in Industrial Anomaly Detection

amazon-research/patchcore-inspection • • CVPR 2022

Being able to spot defective parts is a critical component in large-scale industrial manufacturing.

Paper
Code

DeepWalk: Online Learning of Social Representations

PaddlePaddle/PaddleRec • • 26 Mar 2014

We present DeepWalk, a novel approach for learning latent representations of vertices in a network.

Paper
Code

A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks

hendrycks/error-detection • • 7 Oct 2016

We consider the two related problems of detecting if an example is misclassified or out-of-distribution.

Paper
Code

Is Space-Time Attention All You Need for Video Understanding?

facebookresearch/TimeSformer • • 9 Feb 2021

We present a convolution-free approach to video classification built exclusively on self-attention over space and time.

Paper
Code

MURA: Large Dataset for Abnormality Detection in Musculoskeletal Radiographs

pyaf/DenseNet-MURA-PyTorch • • 11 Dec 2017

To evaluate models robustly and to get an estimate of radiologist performance, we collect additional labels from six board-certified Stanford radiologists on the test set, consisting of 207 musculoskeletal studies.

Paper
Code

LSTM-based Encoder-Decoder for Multi-sensor Anomaly Detection

chickenbestlover/RNN-Time-series-Anomaly-Detection • • 1 Jul 2016

Mechanical devices such as engines, vehicles, aircrafts, etc., are typically instrumented with numerous sensors to capture the behavior and health of the machine.

Paper
Code

Anomaly Detection

Benchmarks Add a Result

Libraries

Datasets

Subtasks

Most implemented papers

Content

Benchmarks

Add a Result