Anomaly localization by modeling perceptual features
arXiv:2008.05369
Abstract
Although unsupervised generative modeling of an image dataset using a Variational AutoEncoder (VAE) has been used to detect anomalous images, or anomalous regions in images, recent works have shown that this method often identifies images or regions that do not concur with human perception, even questioning the usability of generative models for robust anomaly detection. Here, we argue that those issues can emerge from having a simplistic model of the anomaly distribution and we propose a new VAE-based model expressing a more complex anomaly model that is also closer to human perception. This Feature-Augmented VAE is trained by not only reconstructing the input image in pixel space, but also in several different feature spaces, which are computed by a convolutional neural network trained beforehand on a large image dataset. It achieves clear improvement over state-of-the-art methods on the MVTec anomaly detection and localization datasets.
References in corpus (9)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- NIPS 2016 Tutorial: Generative Adversarial Networks
- Uninformed Students: Student-Teacher Anomaly Detection with Discriminative Latent Embeddings
- Improving Unsupervised Defect Segmentation by Applying Structural Similarity to Autoencoders
- WAIC, but Why? Generative Ensembles for Robust Anomaly Detection
- Diagnosing and Enhancing VAE Models
- Detecting Out-of-Distribution Inputs to Deep Generative Models Using Typicality
- Iterative energy-based projection on a normal data manifold for anomaly localization
- Are pre-trained CNNs good feature extractors for anomaly detection in surveillance videos?