A Hierarchical Transformation-Discriminating Generative Model for Few Shot Anomaly Detection

Anomaly detection, the task of identifying unusual samples in data, often relies on a large set of training samples. In this work, we consider the setting of few-shot anomaly detection in images, where only a few images are given at training. We devise a hierarchical generative model that captures the multi-scale patch distribution of each training image. We further enhance the representation of our model by using image transformations and optimize scale-specific patch-discriminators to distinguish between real and fake patches of the image, as well as between different transformations applied to those patches. The anomaly score is obtained by aggregating the patch-based votes of the correct transformation across scales and image regions. We demonstrate the superiority of our method on both the one-shot and few-shot settings, on the datasets of Paris, CIFAR10, MNIST and FashionMNIST as well as in the setting of defect detection on MVTec. In all cases, our method outperforms the recent baseline methods.

[1]  L. Wolf,et al.  Hierarchical Patch VAE-GAN: Generating Diverse Videos from a Single Sample , 2020, NeurIPS.

[2]  Michal Irani,et al.  InGAN: Capturing and Retargeting the “DNA” of a Natural Image , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[3]  Ling Chen,et al.  Learning Representations of Ultrahigh-dimensional Data for Random Distance-based Outlier Detection , 2018, KDD.

[4]  Huidong Jin,et al.  Deep Weakly-supervised Anomaly Detection , 2019 .

[5]  Yu Cheng,et al.  Deep Structured Energy Based Models for Anomaly Detection , 2016, ICML.

[6]  Dani Lischinski,et al.  Non-stationary texture synthesis by adversarial expansion , 2018, ACM Trans. Graph..

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

[8]  Matthew Turk,et al.  CLEAR: Cumulative LEARning for One-Shot One-Class Image Recognition , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[9]  Li Fei-Fei.  Knowledge transfer in learning to recognize visual objects classes , 2006 .

[10]  拓海 杉山,et al.  “Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks”の学習報告 , 2017 .

[11]  Harsha Vardhan Simhadri,et al.  DROCC: Deep Robust One-Class Classification , 2020, ICML.

[12]  Song Han,et al.  Differentiable Augmentation for Data-Efficient GAN Training , 2020, NeurIPS.

[13]  Chuan Li,et al.  Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial Networks , 2016, ECCV.

[14]  Yue Wang,et al.  Rethinking Few-Shot Image Classification: a Good Embedding Is All You Need? , 2020, ECCV.

[15]  Yedid Hoshen,et al.  Classification-Based Anomaly Detection for General Data , 2020, ICLR.

[16]  Charu C. Aggarwal,et al.  Outlier Analysis , 2013, Springer New York.

[17]  VARUN CHANDOLA,et al.  Anomaly detection: A survey , 2009, CSUR.

[18]  Longbing Cao,et al.  Deep Reinforcement Learning for Unknown Anomaly Detection , 2020, ArXiv.

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

[20]  Anton van den Hengel,et al.  Deep Anomaly Detection with Deviation Networks , 2019, KDD.

[21]  M. M. Moya,et al.  One-class classifier networks for target recognition applications , 1993 .

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

[23]  Sungzoon Cho,et al.  Variational Autoencoder based Anomaly Detection using Reconstruction Probability , 2015 .

[24]  Pietro Perona,et al.  One-shot learning of object categories , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[25]  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).

[26]  Roland Vollgraf,et al.  Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms , 2017, ArXiv.

[27]  Wojciech Zaremba,et al.  Improved Techniques for Training GANs , 2016, NIPS.

[28]  Volker Tresp,et al.  Few-Shot One-Class Classification via Meta-Learning , 2019, AAAI.

[29]  James T. Kwok,et al.  Generalizing from a Few Examples , 2019, ACM Comput. Surv..

[30]  Michael Isard,et al.  Lost in quantization: Improving particular object retrieval in large scale image databases , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[31]  Tero Karras,et al.  Training Generative Adversarial Networks with Limited Data , 2020, NeurIPS.

[32]  Anna Kruspe,et al.  One-Way Prototypical Networks , 2019, ArXiv.

[33]  Thomas G. Dietterich,et al.  A Unifying Review of Deep and Shallow Anomaly Detection , 2020, Proceedings of the IEEE.

[34]  Bernhard Schölkopf,et al.  Support Vector Method for Novelty Detection , 1999, NIPS.

[35]  Ran El-Yaniv,et al.  Deep Anomaly Detection Using Geometric Transformations , 2018, NeurIPS.

[36]  Soumith Chintala,et al.  Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks , 2015, ICLR.

[37]  Yu-Chiang Frank Wang,et al.  A Closer Look at Few-shot Classification , 2019, ICLR.

[38]  Tali Dekel,et al.  SinGAN: Learning a Generative Model From a Single Natural Image , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[39]  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).

[40]  Paul W. Fieguth,et al.  A review on computer vision based defect detection and condition assessment of concrete and asphalt civil infrastructure , 2015, Adv. Eng. Informatics.

[41]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[42]  Yee Whye Teh,et al.  Do Deep Generative Models Know What They Don't Know? , 2018, ICLR.