PEDENet: Image Anomaly Localization via Patch Embedding and Density Estimation

A neural network targeting at unsupervised image anomaly localization, called the PEDENet, is proposed in this work. PEDENet contains a patch embedding (PE) network, a density estimation (DE) network, and an auxiliary network called the location prediction (LP) network. The PE network takes local image patches as input and performs dimension reduction to get low-dimensional patch embeddings via a deep encoder structure. Being inspired by the Gaussian Mixture Model (GMM), the DE network takes those patch embeddings, and then predicts the cluster membership of an embedded patch. The sum of membership probabilities is used as a loss term to guide the learning process. The LP network is a Multi-layer Perception (MLP), which takes embeddings from two neighboring patches as input and predicts their relative location. The performance of the proposed PEDENet is evaluated extensively and benchmarked with that of state-of-the-art methods.

[1]  Jian Sun,et al.  Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification , 2015, 2015 IEEE International Conference on Computer Vision (ICCV).

[2]  Jack Beuth,et al.  Anomaly Detection and Classification in a Laser Powder Bed Additive Manufacturing Process using a Trained Computer Vision Algorithm , 2018 .

[3]  Yong Liu,et al.  AnomalyNet: An Anomaly Detection Network for Video Surveillance , 2019, IEEE Transactions on Information Forensics and Security.

[4]  Dorit Merhof,et al.  Modeling the Distribution of Normal Data in Pre-Trained Deep Features for Anomaly Detection , 2021, 2020 25th International Conference on Pattern Recognition (ICPR).

[5]  Hamid R. Rabiee,et al.  Puzzle-AE: Novelty Detection in Images through Solving Puzzles , 2020, ArXiv.

[6]  Bo Zong,et al.  Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection , 2018, ICLR.

[7]  Thomas S. Huang,et al.  One-class SVM for learning in image retrieval , 2001, Proceedings 2001 International Conference on Image Processing (Cat. No.01CH37205).

[8]  Georg Langs,et al.  Unsupervised Anomaly Detection with Generative Adversarial Networks to Guide Marker Discovery , 2017, IPMI.

[9]  Yedid Hoshen,et al.  Transformer-Based Anomaly Segmentation , 2020 .

[10]  Carsten Steger,et al.  Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoders , 2018, VISIGRAPP.

[11]  Venkatesh Saligrama,et al.  Video anomaly detection based on local statistical aggregates , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[12]  Moncef Gabbouj,et al.  AnomalyHop: An SSL-based Image Anomaly Localization Method , 2021, 2021 International Conference on Visual Communications and Image Processing (VCIP).

[13]  Xinfeng Zhang,et al.  A Data-centric Approach to Unsupervised Texture Segmentation Using Principle Representative Patterns , 2019, ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[14]  C.-C. Jay Kuo,et al.  Texture analysis and classification with tree-structured wavelet transform , 1993, IEEE Trans. Image Process..

[15]  Zhenyu Li,et al.  Superpixel Masking and Inpainting for Self-Supervised Anomaly Detection , 2020, BMVC.

[16]  Carsten Steger,et al.  MVTec AD — A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection , 2019, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[17]  Abhijit Mahalanobis,et al.  Attention Guided Anomaly Localization in Images , 2020, ECCV.

[18]  Bin Wang,et al.  Dynamic Texture Synthesis By Incorporating Long-range Spatial and Temporal Correlations , 2021, 2021 International Symposium on Signals, Circuits and Systems (ISSCS).

[19]  Matej Kristan,et al.  Reconstruction by inpainting for visual anomaly detection , 2020, Pattern Recognit..

[20]  Chandan Srivastava,et al.  Support Vector Data Description , 2011 .

[21]  Toby P. Breckon,et al.  GANomaly: Semi-Supervised Anomaly Detection via Adversarial Training , 2018, ACCV.

[22]  Marius Kloft,et al.  Explainable Deep One-Class Classification , 2020, ICLR.

[23]  Alexander Binder,et al.  Deep Semi-Supervised Anomaly Detection , 2019, ICLR.

[24]  Cewu Lu,et al.  Attribute Restoration Framework for Anomaly Detection , 2019, IEEE Transactions on Multimedia.

[25]  Sungroh Yoon,et al.  Patch SVDD: Patch-level SVDD for Anomaly Detection and Segmentation , 2020, ArXiv.

[26]  Bodo Rosenhahn,et al.  Same Same But DifferNet: Semi-Supervised Defect Detection with Normalizing Flows , 2020, 2021 IEEE Winter Conference on Applications of Computer Vision (WACV).

[27]  C. Steger,et al.  Uninformed Students: Student-Teacher Anomaly Detection With Discriminative Latent Embeddings , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[28]  Georg Langs,et al.  f‐AnoGAN: Fast unsupervised anomaly detection with generative adversarial networks , 2019, Medical Image Anal..

[29]  Alexander Binder,et al.  Deep One-Class Classification , 2018, ICML.

[30]  Bir Bhanu,et al.  Towards Visually Explaining Variational Autoencoders , 2020, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[31]  Ye Wang,et al.  Texture Analysis Via Hierarchical Spatial-Spectral Correlation (HSSC) , 2019, 2019 IEEE International Conference on Image Processing (ICIP).