CutPaste: Self-Supervised Learning for Anomaly Detection and Localization

@article{Li2021CutPasteSL,
  title={CutPaste: Self-Supervised Learning for Anomaly Detection and Localization},
  author={Chun-Liang Li and Kihyuk Sohn and Jinsung Yoon and Tomas Pfister},
  journal={2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2021},
  pages={9659-9669}
}
We aim at constructing a high performance model for defect detection that detects unknown anomalous patterns of an image without anomalous data. To this end, we propose a two-stage framework for building anomaly detectors using normal training data only. We first learn self-supervised deep representations and then build a generative one-class classifier on learned representations. We learn representations by classifying normal data from the CutPaste, a simple data augmentation strategy that… 
[PDF] Semantic Reader
162 Citations
Highly Influential Citations
50
Background Citations
86
Methods Citations
97
Results Citations
7

162 Citations

Just Noticeable Learning for Unsupervised Anomaly Localization and Detection

  • Ying Zhao
  • Computer Science
    2022 IEEE International Conference on Multimedia and Expo (ICME)
  • 2022
A Just Noticeable learning for unsupervised anomaly Localization and Detection (JNLD) that embeds just noticeable learning in the sub-networks of reconstruction and segmentation to di-rectly localize the defects without any complex additional post-processing.

CRADL: Contrastive Representations for Unsupervised Anomaly Detection and Localization

This work proposes CRADL whose core idea is to model the distribution of normal samples directly in the low-dimensional representation space of an encoder trained with a contrastive pretext-task and shows competitive or superior performance to state-of-the-art methods.

Transfer Learning Gaussian Anomaly Detection by Fine-Tuning Representations

A new method to overcome catastrophic forgetting and thus successfully tune pre-trained representations for AD in the transfer learning setting is proposed and ablation studies demonstrate the importance of the induced Gaussian distribution as well as the robustness of the proposed fine-tuning scheme with respect to the choice of augmentations.

Self-Supervised Predictive Convolutional Attentive Block for Anomaly Detection

The proposed self-supervised block is generic and can easily be incorporated into various state-of-the-art anomaly detection methods, providing empirical evidence that shows considerable performance improvements on MVTec AD, Avenue, and ShanghaiTech.

Semi-Supervised Anomaly Detection with Contrastive Regularization

This work introduces a new deep anomaly detector enforcing an anomaly distance constraint on the norm of the representations while using contrastive learning on the direction of the features to improve the robustness of any distance-based anomaly detector.

Self-Trained One-class Classification for Unsupervised Anomaly Detection

This work focuses on unsupervised AD problems whose entire training data are unlabeled and may contain both normal and anomalous samples, and proposes a self-training of deep representation one-class classifiers (STOC) that iteratively refines the data and deep representations.

Self-supervise, Refine, Repeat: Improving Unsupervised Anomaly Detection

This paper focuses on fully unsupervised AD, in which the entire training dataset, containing both normal and anomalous samples, is unlabeled, and proposes to improve the robustness of one-class classification trained on self-supervised representations using a data re finement process.

No Shifted Augmentations (NSA): compact distributions for robust self-supervised Anomaly Detection

This work investigates how the geometrical compactness of the ID feature distribution makes isolating and detecting outliers easier, especially in the realistic situation when ID training data is polluted, and proposes novel architectural modifications to the self-supervised feature learning step, that enable such compact distributions for ID data to be learned.

Self-Supervised Masking for Unsupervised Anomaly Detection and Localization

A self-supervised learning approach through random masking and then restoring, named Self-Supervised Masking (SSM) for unsupervised anomaly detection and localization, which outperforms several state-of-the-arts for both anomalies detection and anomaly localization.

Self-Supervised Anomaly Detection via Neural Autoregressive Flows with Active Learning

This work proposes a novel active learning (AL) scheme that relied on neural autoregressive flows (NAF) for self-supervised anomaly detection, specifically on small-scale data, which outperforms existing baselines on multiple time series and tabular datasets and a real-world application in advanced manufacturing.
...

References

SHOWING 1-10 OF 65 REFERENCES

MVTec AD — A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection

This work introduces the MVTec Anomaly Detection (MVTec AD) dataset containing 5354 high-resolution color images of different object and texture categories, and conducts a thorough evaluation of current state-of-the-art unsupervised anomaly detection methods based on deep architectures such as convolutional autoencoders, generative adversarial networks, and feature descriptors using pre-trained convolved neural networks.

Deep Anomaly Detection with Outlier Exposure

In extensive experiments on natural language processing and small- and large-scale vision tasks, it is found that Outlier Exposure significantly improves detection performance and that cutting-edge generative models trained on CIFar-10 may assign higher likelihoods to SVHN images than to CIFAR-10 images; OE is used to mitigate this issue.

Modeling the Distribution of Normal Data in Pre-Trained Deep Features for Anomaly Detection

This work demonstrates that deep feature representations learned by discriminative models on large natural image datasets are well suited to describe normality and detect even subtle anomalies in a transfer learning setting and finds that the principal components containing little variance in normal data are crucial for discriminating between normal and anomalous instances.

Uninformed Students: Student-Teacher Anomaly Detection With Discriminative Latent Embeddings

A powerful student-teacher framework for the challenging problem of unsupervised anomaly detection and pixel-precise anomaly segmentation in high-resolution images by trained to regress the output of a descriptive teacher network that was pretrained on a large dataset of patches from natural images.

Explainable Deep One-Class Classification

An explainable deep one-class classification method, Fully Convolutional Data Description (FCDD), where the mapped samples are themselves also an explanation heatmap, which yields competitive detection performance and provides reasonable explanations on common anomaly detection benchmarks with CIFAR-10 and ImageNet.

GANomaly: Semi-Supervised Anomaly Detection via Adversarial Training

This work introduces a novel anomaly detection model, by using a conditional generative adversarial network that jointly learns the generation of high-dimensional image space and the inference of latent space and shows the model efficacy and superiority over previous state-of-the-art approaches.

Improved anomaly detection by training an autoencoder with skip connections on images corrupted with Stain-shaped noise

This work proposes to corrupt clean images with a synthetic noise model to prevent the convergence of the network towards the identity mapping, and introduces an original Stain noise model for that purpose, which favors the reconstruction of clean images from arbitrary real-world images, regardless of the actual defects appearance.

Deep One-Class Classification

This paper introduces a new anomaly detection method—Deep Support Vector Data Description—, which is trained on an anomaly detection based objective and shows the effectiveness of the method on MNIST and CIFAR-10 image benchmark datasets as well as on the detection of adversarial examples of GTSRB stop signs.

A Unifying Review of Deep and Shallow Anomaly Detection

This review aims to identify the common underlying principles and the assumptions that are often made implicitly by various methods in deep learning, and draws connections between classic “shallow” and novel deep approaches and shows how this relation might cross-fertilize or extend both directions.

Inverse-Transform AutoEncoder for Anomaly Detection

A selected set of transformations based on human priors is used to erase certain targeted information from input data using an inverse-transform autoencoder to embed corresponding erased information during the restoration of the original data.
...