Detecting Anomalous Faces with 'No Peeking' Autoencoders

Detecting anomalous faces has important applications. For example, a system might tell when a train driver is incapacitated by a medical event, and assist in adopting a safe recovery strategy. These applications are demanding, because they require accurate detection of rare anomalies that may be seen only at runtime. Such a setting causes supervised methods to perform poorly. We describe a method for detecting an anomalous face image that meets these requirements. We construct a feature vector that reliably has large entries for anomalous images, then use various simple unsupervised methods to score the image based on the feature. Obvious constructions (autoencoder codes; autoencoder residuals) are defeated by a 'peeking' behavior in autoencoders. Our feature construction removes rectangular patches from the image, predicts the likely content of the patch conditioned on the rest of the image using a specially trained autoencoder, then compares the result to the image. High scores suggest that the patch was difficult for an autoencoder to predict, and so is likely anomalous. We demonstrate that our method can identify real anomalous face images in pools of typical images, taken from celeb-A, that is much larger than usual in state-of-the-art experiments. A control experiment based on our method with another set of normal celebrity images - a 'typical set', but nonceleb-A are not identified as anomalous; confirms this is not due to special properties of celeb-A.

[1]  H.Y.K. Lau,et al.  A real-time computer vision system for detecting defects in textile fabrics , 2005, 2005 IEEE International Conference on Industrial Technology.

[2]  Sabine Süsstrunk,et al.  Outlier Modeling in Image Matching , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[3]  Trevor Hastie,et al.  Regularization Paths for Generalized Linear Models via Coordinate Descent. , 2010, Journal of statistical software.

[4]  Alexei A. Efros,et al.  Context Encoders: Feature Learning by Inpainting , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[5]  Alexander J. Smola,et al.  Deep Sets , 2017, 1703.06114.

[6]  Pascal Vincent,et al.  Stacked Denoising Autoencoders: Learning Useful Representations in a Deep Network with a Local Denoising Criterion , 2010, J. Mach. Learn. Res..

[7]  Paul A. Viola,et al.  Rapid object detection using a boosted cascade of simple features , 2001, Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. CVPR 2001.

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

[9]  Geoffrey E. Hinton,et al.  Autoencoders, Minimum Description Length and Helmholtz Free Energy , 1993, NIPS.

[10]  Omkar M. Parkhi,et al.  VGGFace2: A Dataset for Recognising Faces across Pose and Age , 2017, 2018 13th IEEE International Conference on Automatic Face & Gesture Recognition (FG 2018).

[11]  P. Mahalanobis On the generalized distance in statistics , 1936 .

[12]  B. Efron Size, power and false discovery rates , 2007, 0710.2245.

[13]  Yoshua Bengio,et al.  Generative Adversarial Networks , 2014, ArXiv.

[14]  Geoffrey E. Hinton,et al.  Reducing the Dimensionality of Data with Neural Networks , 2006, Science.

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

[16]  A. Ertuzun,et al.  Defect detection in textile fabric images using wavelet transforms and independent component analysis , 2006, Pattern Recognition and Image Analysis.

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

[18]  Shachar Fleishman,et al.  Novelty Detection with GAN , 2018, ArXiv.

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

[20]  Honglak Lee,et al.  Attribute2Image: Conditional Image Generation from Visual Attributes , 2015, ECCV.

[21]  William J. Christmas,et al.  An anomaly detection approach to face spoofing detection: A new formulation and evaluation protocol , 2017, 2017 IEEE International Joint Conference on Biometrics (IJCB).

[22]  Max Welling,et al.  Auto-Encoding Variational Bayes , 2013, ICLR.

[23]  Xiaogang Wang,et al.  Deep Learning Face Attributes in the Wild , 2014, 2015 IEEE International Conference on Computer Vision (ICCV).

[24]  Pavel Kisilev,et al.  Unsupervised detection of abnormalities in medical images using salient features , 2014, Medical Imaging.

[25]  Aditya Deshpande,et al.  Learning Diverse Image Colorization , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[26]  L Sirovich,et al.  Low-dimensional procedure for the characterization of human faces. , 1987, Journal of the Optical Society of America. A, Optics and image science.

[27]  Volker Tresp,et al.  Semi-Supervised Outlier Detection Using a Generative and Adversary Framework , 2018 .

[28]  Guillaume Lample,et al.  Fader Networks: Manipulating Images by Sliding Attributes , 2017, NIPS.

[29]  David Berthelot,et al.  BEGAN: Boundary Equilibrium Generative Adversarial Networks , 2017, ArXiv.