Multi-head enhanced self-attention network for novelty detection

Abstract One-class classification (OCC) is a classical problem in computer vision that can be described as the task of classifying outlier class samples (OC samples) from the OCC model trained on inlier class samples (IC samples) when datasets are highly biased toward one class due to the insufficient sample size of the other class. Currently, the adversarial learning OCC (ALOCC) method has been proven to significantly improve OCC performance. However, its drawbacks include instability issues and non-evident reconstruction between the IC and OC samples. Therefore, we propose multihead enhanced self-attention in the ALOCC network, thereby increasing the difference between the IC and OC samples and significantly increasing OCC accuracy compared with ALOCC accuracy. For training, we propose a new loss, called adversarial-balance loss, that effectively solves the training instability problem, further increasing OCC accuracy. The experiments show the effectiveness of the proposed method compared with state-of-art methods.

[1]  Angshul Majumdar,et al.  RODEO: Robust DE-aliasing autoencOder for real-time medical image reconstruction , 2017, Pattern Recognit..

[2]  Guido Sanguinetti,et al.  Information theoretic novelty detection , 2010, Pattern Recognit..

[3]  Abhinav Gupta,et al.  Non-local Neural Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[4]  Léon Bottou,et al.  Wasserstein GAN , 2017, ArXiv.

[5]  Mirella Lapata,et al.  Long Short-Term Memory-Networks for Machine Reading , 2016, EMNLP.

[6]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[7]  Nicu Sebe,et al.  Training Adversarial Discriminators for Cross-Channel Abnormal Event Detection in Crowds , 2017, 2019 IEEE Winter Conference on Applications of Computer Vision (WACV).

[8]  Mahmood Fathy,et al.  Video anomaly detection and localisation based on the sparsity and reconstruction error of auto-encoder , 2016 .

[9]  Nicu Sebe,et al.  Learning Deep Representations of Appearance and Motion for Anomalous Event Detection , 2015, BMVC.

[10]  Aaron C. Courville,et al.  Improved Training of Wasserstein GANs , 2017, NIPS.

[11]  Alan C. Bovik,et al.  Image information and visual quality , 2006, IEEE Trans. Image Process..

[12]  Md. Rafiqul Islam,et al.  A survey of anomaly detection techniques in financial domain , 2016, Future Gener. Comput. Syst..

[13]  Michal Irani,et al.  Detecting Irregularities in Images and in Video , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[14]  B. Ravi Kiran,et al.  An overview of deep learning based methods for unsupervised and semi-supervised anomaly detection in videos , 2018, J. Imaging.

[15]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[16]  Ashish Vaswani,et al.  Self-Attention with Relative Position Representations , 2018, NAACL.

[17]  Nuno Vasconcelos,et al.  Anomaly Detection and Localization in Crowded Scenes , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Amparo Alonso-Betanzos,et al.  Fault detection via recurrence time statistics and one-class classification , 2016, Pattern Recognit. Lett..

[19]  Witold Pedrycz,et al.  Dual autoencoders features for imbalance classification problem , 2016, Pattern Recognit..

[20]  Daniel P. Robinson,et al.  Provable Self-Representation Based Outlier Detection in a Union of Subspaces , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[21]  Xiaogang Wang,et al.  Residual Attention Network for Image Classification , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[22]  Constantine Caramanis,et al.  Robust PCA via Outlier Pursuit , 2010, IEEE Transactions on Information Theory.

[23]  Lianwen Jin,et al.  A Multi-Object Rectified Attention Network for Scene Text Recognition , 2019, Pattern Recognit..

[24]  Sameer Singh,et al.  An approach to novelty detection applied to the classification of image regions , 2004, IEEE Transactions on Knowledge and Data Engineering.

[25]  Mahmood Fathy,et al.  Deep-Cascade: Cascading 3D Deep Neural Networks for Fast Anomaly Detection and Localization in Crowded Scenes , 2017, IEEE Transactions on Image Processing.

[26]  Nuno Vasconcelos,et al.  Anomaly detection in crowded scenes , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

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

[28]  René Vidal,et al.  Dual Principal Component Pursuit , 2015, 2015 IEEE International Conference on Computer Vision Workshop (ICCVW).

[29]  Nelson D. A. Mascarenhas,et al.  Ultrasound Image Despeckling Using Stochastic Distance-Based BM3D , 2017, IEEE Transactions on Image Processing.

[30]  Joel A. Tropp,et al.  Robust Computation of Linear Models by Convex Relaxation , 2012, Foundations of Computational Mathematics.

[31]  Maoguo Gong,et al.  Structured self-attention architecture for graph-level representation learning , 2020, Pattern Recognit..

[32]  Mahmood Fathy,et al.  Adversarially Learned One-Class Classifier for Novelty Detection , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[33]  Kristen Grauman,et al.  Observe locally, infer globally: A space-time MRF for detecting abnormal activities with incremental updates , 2009, CVPR.

[34]  Yang Yu,et al.  Unsupervised Representation Learning with Deep Convolutional Neural Network for Remote Sensing Images , 2017, ICIG.

[35]  Heitor Silvério Lopes,et al.  A study of deep convolutional auto-encoders for anomaly detection in videos , 2018, Pattern Recognit. Lett..

[36]  Eduardo Fidalgo,et al.  Boosting image classification through semantic attention filtering strategies , 2018, Pattern Recognit. Lett..