Attribute Restoration Framework for Anomaly Detection

Abstract

With the recent advances in deep neural networks, anomaly detection in multimedia has received much attention in the computer vision community. While reconstruction-based methods have recently shown great promise for anomaly detection, the information equivalence among input and supervision for reconstruction tasks can not effectively force the network to learn semantic feature embeddings. We here propose to break this equivalence by erasing selected attributes from the original data and reformulate it as a restoration task, where the normal and the anomalous data are expected to be distinguishable based on restoration errors. Through forcing the network to restore the original image, the semantic feature embeddings related to the erased attributes are learned by the network. During testing phases, because anomalous data are restored with the attribute learned from the normal data, the restoration error is expected to be large. Extensive experiments have demonstrated that the proposed method significantly outperforms several state-of-the-arts on multiple benchmark datasets, especially on ImageNet, increasing the AUROC of the top-performing baseline by 10.1%. We also evaluate our method on a real-world anomaly detection dataset MVTec AD and a video anomaly detection dataset ShanghaiTech.

References in corpus (12)

• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Deep Learning for Anomaly Detection: A Survey
• Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddings
• Autoencoding beyond pixels using a learned similarity metric
• Efficient GAN-Based Anomaly Detection
• Deep Anomaly Detection with Outlier Exposure
• Deep Anomaly Detection Using Geometric Transformations
• adVAE: A self-adversarial variational autoencoder with Gaussian anomaly prior knowledge for anomaly detection
• Coherence Pursuit: Fast, Simple, and Robust Principal Component Analysis
• Semi-Supervised Learning with Context-Conditional Generative Adversarial Networks
• Benchmarking Neural Network Robustness to Common Corruptions and Surface Variations
• Self-Supervised Feature Learning by Learning to Spot Artifacts

Cited by in corpus (12)

• Sub-Image Anomaly Detection with Deep Pyramid Correspondences
• Student-Teacher Feature Pyramid Matching for Anomaly Detection
• Learning and Evaluating Representations for Deep One-class Classification
• Self-Supervised Anomaly Detection in Computer Vision and Beyond: A Survey and Outlook
• Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentation
• Transfer Learning Gaussian Anomaly Detection by Fine-tuning Representations
• Improving Novelty Detection using the Reconstructions of Nearest Neighbours
• ESAD: End-to-end Deep Semi-supervised Anomaly Detection
• Fine-grained Anomaly Detection via Multi-task Self-Supervision
• Data augmentation and pre-trained networks for extremely low data regimes unsupervised visual inspection
• P-WAE: Generalized Patch-Wasserstein Autoencoder for Anomaly Screening
• Deep Unsupervised Image Anomaly Detection: An Information Theoretic Framework