Deep One-Class Classification via Interpolated Gaussian Descriptor

  title={Deep One-Class Classification via Interpolated Gaussian Descriptor},
  author={Yuanhong Chen and Yu Tian and Guansong Pang and G. Carneiro},
One-class classification (OCC) aims to learn an effective data description to enclose all normal training samples and detect anomalies based on the deviation from the data description. Current state-of-the-art OCC models learn a compact normality description by hyper-sphere minimisation, but they often suffer from overfitting the training data, especially when the training set is small or contaminated with anomalous samples. To address this issue, we introduce the interpolated Gaussian… 

Figures and Tables from this paper

Catching Both Gray and Black Swans: Open-set Supervised Anomaly Detection
This paper proposes a novel approach that learns disentangled representations of abnormalities illustrated by seen anomalies, pseudo anomalies, and latent residual anomalies (i.e., samples that have unusual residuals compared to the normal data in a latent space), with the last two abnormalities designed to detect unseen anomalies.
Perturbation Learning Based Anomaly Detection
Compared with the state-of-the-art of anomaly detection, this method does not require any assumption about the shape of the decision boundary and has fewer hyper-parameters to determine.
Pixel-wise Energy-biased Abstention Learning for Anomaly Segmentation on Complex Urban Driving Scenes
A new anomaly segmentation method that explores pixel-wise energy-biased abstention learning (AL) with a model that learns an adaptive pixel-level anomaly class, and an energy-based model (EBM) that learns inlier pixel distribution is proposed.
A Survey on Unsupervised Industrial Anomaly Detection Algorithms
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.
Unsupervised Anomaly Detection in Medical Images with a Memory-augmented Multi-level Cross-attentional Masked Autoencoder
A new reconstruction-based UAD approach that addresses the low-reconstruction error issue for anomalous images and achieves SOTA anomaly detection and localisation on colonoscopy and Covid-19 Chest X-ray datasets.
Contrastive Transformer-based Multiple Instance Learning for Weakly Supervised Polyp Frame Detection
A novel convolutional transformer-based multiple instance learning method designed to identify abnormal frames from anomalous videos, which achieves a detection accuracy that is substantially better than current state-of-the-art approaches on a new large-scale colonoscopy video dataset introduced in this work.
Anatomy-aware Self-supervised Learning for Anomaly Detection in Chest Radiographs
This is the first SSL model to employ anatomical information as a pretext task and exhibit area under curves (AUC) of 92.1%, 78.7%, and 81.9%, which are the highest among existing UAD models.
CTFSH: Full Head CT Anomaly Detection with Unsupervised Learning
This work’s aim is to build a comparison between two commonly used baselines for UAD of volumetric CT scans, and is the first work to show a comparison regarding the usage of skull-stripping.


Adversarially Learned One-Class Classifier for Novelty Detection
The results on MNIST and Caltech-256 image datasets, along with the challenging UCSD Ped2 dataset for video anomaly detection illustrate that the proposed method learns the target class effectively and is superior to the baseline and state-of-the-art methods.
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.
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.
OCGAN: One-Class Novelty Detection Using GANs With Constrained Latent Representations
A novel model called OCGAN is presented for the classical problem of one-class novelty detection, where, given a set of examples from a particular class, the goal is to determine if a query example is from the same class using a de-noising auto-encoder network.
Learning and Evaluating Representations for Deep One-class Classification
A novel distribution-augmented contrastive learning that extends training distributions via data augmentation to obstruct the uniformity of contrastive representations and argues that classifiers inspired by the statistical perspective in generative or discriminative models are more effective than existing approaches.
Learning Semantic Context from Normal Samples for Unsupervised Anomaly Detection
This work presents a Semantic Context based Anomaly Detection Network, SCADN, for unsupervised anomaly detection by learning the semantic context from the normal samples by generating multi-scale striped masks and training a generative adversarial network to reconstruct the unseen regions.
Skip-GANomaly: Skip Connected and Adversarially Trained Encoder-Decoder Anomaly Detection
This work introduces an unsupervised anomaly detection model, trained only on the normal (non-anomalous, plentiful) samples in order to learn the normality distribution of the domain, and hence detect abnormality based on deviation from this model.
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.
CutPaste: Self-Supervised Learning for Anomaly Detection and Localization
This work proposes a two-stage framework for building anomaly detectors using normal training data only, which first learns self-supervised deep representations and then builds a generative one-class classifier on learned representations.
Learning Deep Features for One-Class Classification
A novel deep-learning-based approach for one-class transfer learning in which labeled data from an unrelated task is used for feature learning in one- class classification and achieves significant improvements over the state-of-the-art.