Deep Anomaly Detection Using Geometric Transformations

Abstract

We consider the problem of anomaly detection in images, and present a new detection technique. Given a sample of images, all known to belong to a "normal" class (e.g., dogs), we show how to train a deep neural model that can detect out-of-distribution images (i.e., non-dog objects). The main idea behind our scheme is to train a multi-class model to discriminate between dozens of geometric transformations applied on all the given images. The auxiliary expertise learned by the model generates feature detectors that effectively identify, at test time, anomalous images based on the softmax activation statistics of the model when applied on transformed images. We present extensive experiments using the proposed detector, which indicate that our algorithm improves state-of-the-art methods by a wide margin.

References in corpus (9)

• Adam: A Method for Stochastic Optimization
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Wide Residual Networks
• Understanding Neural Networks Through Deep Visualization
• Deep Structured Energy Based Models for Anomaly Detection
• Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery
• Deep Active Learning over the Long Tail
• Model Selection for Anomaly Detection

Cited by in corpus (79)

• Self-Supervised Representation Learning: Introduction, Advances and Challenges
• Using Self-Supervised Learning Can Improve Model Robustness and Uncertainty
• Sub-Image Anomaly Detection with Deep Pyramid Correspondences
• Generalized Out-of-Distribution Detection: A Survey
• CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances
• Omni-frequency Channel-selection Representations for Unsupervised Anomaly Detection
• The Fishyscapes Benchmark: Measuring Blind Spots in Semantic Segmentation
• Anomaly Detection in Medical Imaging with Deep Perceptual Autoencoders
• Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detection
• A Survey on Unsupervised Anomaly Detection Algorithms for Industrial Images
• Deep Nearest Neighbor Anomaly Detection
• Student-Teacher Feature Pyramid Matching for Anomaly Detection
• Explainable Deep One-Class Classification
• Self-Supervised Anomaly Detection in Computer Vision and Beyond: A Survey and Outlook
• MOCCA: Multi-Layer One-Class ClassificAtion for Anomaly Detection
• Dual-distribution discrepancy with self-supervised refinement for anomaly detection in medical images
• DASVDD: Deep Autoencoding Support Vector Data Descriptor for Anomaly Detection
• Anomaly Detection in Aerial Videos with Transformers
• Attribute Restoration Framework for Anomaly Detection
• Anomaly Detection based on Zero-Shot Outlier Synthesis and Hierarchical Feature Distillation
• A Survey on Open Set Recognition
• Detecting Outliers with Foreign Patch Interpolation
• Masked Transformer for image Anomaly Localization
• Unsupervised Outlier Detection using Memory and Contrastive Learning
• Understanding Anomaly Detection with Deep Invertible Networks through Hierarchies of Distributions and Features
• Revealing quantum chaos with machine learning
• Unsupervised Learning of Anomaly Detection from Contaminated Image Data using Simultaneous Encoder Training
• An Iterative Method for Unsupervised Robust Anomaly Detection Under Data Contamination
• Deep Semi-Supervised Anomaly Detection
• Robust Subspace Recovery Layer for Unsupervised Anomaly Detection
• Neural Transformation Learning for Deep Anomaly Detection Beyond Images
• Reconstruction from edge image combined with color and gradient difference for industrial surface anomaly detection
• Image/Video Deep Anomaly Detection: A Survey
• FAIR: Frequency-aware Image Restoration for Industrial Visual Anomaly Detection
• A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challenges
• On the Modularity of Hypernetworks
• Transfer Learning Gaussian Anomaly Detection by Fine-tuning Representations
• Visual Anomaly Detection for Images: A Survey
• Understanding Open-Set Recognition by Jacobian Norm and Inter-Class Separation
• Puzzle-AE: Novelty Detection in Images through Solving Puzzles
• Rethinking Assumptions in Deep Anomaly Detection
• Further Analysis of Outlier Detection with Deep Generative Models
• Anomalous Example Detection in Deep Learning: A Survey
• Evaluation of Point Pattern Features for Anomaly Detection of Defect within Random Finite Set Framework
• ESAD: End-to-end Deep Semi-supervised Anomaly Detection
• OneFlow: One-class flow for anomaly detection based on a minimal volume region
• Out-of-domain Detection for Natural Language Understanding in Dialog Systems
• Double-Adversarial Activation Anomaly Detection: Adversarial Autoencoders are Anomaly Generators
• DROCC: Deep Robust One-Class Classification
• Focus Your Distribution: Coarse-to-Fine Non-Contrastive Learning for Anomaly Detection and Localization
• SLA${}^2$P: Self-supervised Anomaly Detection with Adversarial Perturbation
• Constrained Contrastive Distribution Learning for Unsupervised Anomaly Detection and Localisation in Medical Images
• Non-Negative Bregman Divergence Minimization for Deep Direct Density Ratio Estimation
• Discriminative-Generative Representation Learning for One-Class Anomaly Detection
• Fine-grained Anomaly Detection via Multi-task Self-Supervision
• Anomaly Detection on X-Rays Using Self-Supervised Aggregation Learning
• Deep Variational Semi-Supervised Novelty Detection
• Novelty Detection via Robust Variational Autoencoding
• Uncertainty-Aware Deep Classifiers using Generative Models
• Multi-view Deep One-class Classification: A Systematic Exploration
• Self-supervised Pseudo Multi-class Pre-training for Unsupervised Anomaly Detection and Segmentation in Medical Images
• Exploring Deep Anomaly Detection Methods Based on Capsule Net
• Enhancing Unsupervised Anomaly Detection with Score-Guided Network
• Unsupervised Anomaly Detection From Semantic Similarity Scores
• Medical Knowledge-Guided Deep Learning for Imbalanced Medical Image Classification
• Deep Context-Aware Novelty Detection
• Discriminative Multi-level Reconstruction under Compact Latent Space for One-Class Novelty Detection
• Learning Augmentation Distributions using Transformed Risk Minimization
• Anomaly Detection by One Class Latent Regularized Networks
• Deep Residual Flow for Out of Distribution Detection
• OMASGAN: Out-of-Distribution Minimum Anomaly Score GAN for Sample Generation on the Boundary
• Multi-Class Anomaly Detection
• New Perspective on Progressive GANs Distillation for One-class Novelty Detection
• Using Fictitious Class Representations to Boost Discriminative Zero-Shot Learners
• Hierarchical Models: Intrinsic Separability in High Dimensions
• A Generic Machine Learning Framework for Fully-Unsupervised Anomaly Detection with Contaminated Data
• Data augmentation and pre-trained networks for extremely low data regimes unsupervised visual inspection
• P-WAE: Generalized Patch-Wasserstein Autoencoder for Anomaly Screening
• Y-GAN: Learning Dual Data Representations for Efficient Anomaly Detection