• Corpus ID: 235436036

Towards Total Recall in Industrial Anomaly Detection

@article{Roth2021TowardsTR,
  title={Towards Total Recall in Industrial Anomaly Detection},
  author={Karsten Roth and Latha Pemula and Joaquin Zepeda and Bernhard Sch{\"o}lkopf and Thomas Brox and Peter V. Gehler},
  journal={ArXiv},
  year={2021},
  volume={abs/2106.08265}
}
Being able to spot defective parts is a critical component in large-scale industrial manufacturing. A particular chal-lenge that we address in this work is the cold-start problem: fit a model using nominal (non-defective) example images only. While handcrafted solutions per class are possible, the goal is to build systems that work well simultaneously on many different tasks automatically. The best peform-ing approaches combine embeddings from ImageNet models with an outlier detection model. In… 
View With Semantic Reader
21 Citations
Highly Influential Citations
11
Background Citations
13
Methods Citations
11
Results Citations
1

Figures and Tables from this paper

21 Citations

Citation Type

Citation Type

CFA: Coupled-hypersphere-based Feature Adaptation for Target-Oriented Anomaly Localization
TLDR
CFA adopts transfer learning to increase the normal feature density so that abnormal features can be clearly distinguished by applying patch descriptor and memory bank to a pretrained CNN and outperforms the previous methods quantitatively and qualitatively.
PAEDID: Patch Autoencoder Based Deep Image Decomposition For Pixel-level Defective Region Segmentation
TLDR
This work presents an unsupervised patch autoencoder based deep image decomposition (PAEDID) method for defective region segmentation and demonstrates the effectiveness of the PAEDID method in simulation studies and an industrial dataset in the case study.
FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows
TLDR
This work proposes FastFlow implemented with 2D normalizing flows and uses it as the probability distribution estimator for unsupervised anomaly detection and localization and achieves 99.4% AUC in anomaly detection with high inference efficiency.
CAINNFlow: Convolutional block Attention modules and Invertible Neural Networks Flow for anomaly detection and localization tasks
TLDR
CAINNFlow achieves advanced levels of accuracy and inference efficiency based on CNN and Transformer backbone networks as feature extractors, and CAINN Flow achieves a pixel-level AUC of 98 .
ALGAN: Anomaly Detection by Generating Pseudo Anomalous Data via Latent Variables
TLDR
An Anomalous Latent variable Generative Adversarial Network (ALGAN) in which the GAN generator produces pseudo-anomalous data as well as fake-normal data, whereas the discriminator is trained to distinguish between normal and pseudo-AnomalousData.
Benchmarking Unsupervised Anomaly Detection and Localization
TLDR
A comprehensive comparison of existing methods in unsupervised anomaly detection and localization tasks and adds a comparison of inference previously ignored by the community to inspire further research.
MemSeg: A semi-supervised method for image surface defect detection using differences and commonalities
TLDR
In order to fully fuse the memory information with the high-level features of the input image, MemSeg introduces a multi-scale feature fusion module (MSFF Module) and a novel spatial attention module, which greatly improves the model precision of anomaly localization.
A Survey on Unsupervised Industrial Anomaly Detection Algorithms
TLDR
A thorough overview of recently proposed unsupervised algorithms for visual anomaly detection covering categories, whose innovation points and frameworks are described in detail are provided and expected to assist both the research community and industry in developing a broader and cross-domain perspective.
An Empirical Investigation of 3D Anomaly Detection and Segmentation
TLDR
A simple 3D-only method is uncovered that outperforms all recent approaches while not using deep learning, external pretraining datasets, or color information, and is combined with 2D color features.
Anomalib: A Deep Learning Library for Anomaly Detection
TLDR
This paper introduces anomalib1, a novel library for unsupervised anomaly detection and localization that provides algorithms from the literature and a set of tools to design custom anomaly detection algorithms via a plug-andplay approach.
...
...

References

SHOWING 1-10 OF 65 REFERENCES
Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flows
TLDR
DifferNet leverages the descriptiveness of features extracted by convolutional neural networks to estimate their density using normalizing flows to develop a scoring function that indicates defects.
MVTec AD — A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection
TLDR
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.
ImageNet: A large-scale hierarchical image database
TLDR
A new database called “ImageNet” is introduced, a large-scale ontology of images built upon the backbone of the WordNet structure, much larger in scale and diversity and much more accurate than the current image datasets.
Future Frame Prediction for Anomaly Detection - A New Baseline
TLDR
This paper proposes to tackle the anomaly detection problem within a video prediction framework by introducing a motion (temporal) constraint in video prediction by enforcing the optical flow between predicted frames and ground truth frames to be consistent, and is the first work that introduces a temporal constraint into the video prediction task.
PaDiM: a Patch Distribution Modeling Framework for Anomaly Detection and Localization
TLDR
PaDiM outperforms current state-of-the-art approaches for both anomaly detection and localization on the MVTec AD and STC datasets and is a good candidate for many industrial applications.
Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentation
TLDR
Support vector data description is a long-standing algorithm used for an anomaly detection, and its deep learning variant is extended to the patch-based method using self-supervised learning, which enables anomaly segmentation and improves detection performance.
Attention Guided Anomaly Localization in Images
TLDR
This work proposes Convolutional Adversarial Variational autoencoder with Guided Attention (CAVGA), which localizes the anomaly with a convolutional latent variable to preserve the spatial information and outperforms state-of-the-art anomaly localization methods on several datasets.
Uninformed Students: Student-Teacher Anomaly Detection With Discriminative Latent Embeddings
TLDR
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.
Support Vector Data Description
Surface defect saliency of magnetic tile
  • Yibin Huang, C. Qiu, Kui Yuan
  • Computer Science
    2018 IEEE 14th International Conference on Automation Science and Engineering (CASE)
  • 2018
TLDR
It is shown that the model exceeds the state of the art in saliency detection of magnetic tiles, in which it both effectively and explicitly maps multiple surface defects from low-contrast images.
...
...