Adversarially Learned One-Class Classifier for Novelty Detection

TL;DR

This paper introduces an adversarial deep learning framework for one-class classification that effectively detects anomalies and novelties in images and videos by training two competing networks.

Contribution

It proposes an end-to-end adversarial architecture for one-class classification, enhancing inlier and outlier separation without requiring novelty data during training.

Findings

Outperforms baseline and state-of-the-art methods on MNIST, Caltech-256, and UCSD Ped2 datasets.

Effectively learns the target class for anomaly detection.

Demonstrates robustness across image and video applications.

Abstract

Novelty detection is the process of identifying the observation(s) that differ in some respect from the training observations (the target class). In reality, the novelty class is often absent during training, poorly sampled or not well defined. Therefore, one-class classifiers can efficiently model such problems. However, due to the unavailability of data from the novelty class, training an end-to-end deep network is a cumbersome task. In this paper, inspired by the success of generative adversarial networks for training deep models in unsupervised and semi-supervised settings, we propose an end-to-end architecture for one-class classification. Our architecture is composed of two deep networks, each of which trained by competing with each other while collaborating to understand the underlying concept in the target class, and then classify the testing samples. One network works as the novelty detector, while the other supports it by enhancing the inlier samples and distorting the outliers. The intuition is that the separability of the enhanced inliers and distorted outliers is much better than deciding on the original samples. The proposed framework applies to different related applications of anomaly and outlier detection in images and videos. The results on MNIST and Caltech-256 image datasets, along with the challenging UCSD Ped2 dataset for video anomaly detection illustrate that our proposed method learns the target class effectively and is superior to the baseline and state-of-the-art methods.