Learning Deep Features for One-Class Classification

TL;DR

This paper introduces a deep learning approach for one-class classification that learns descriptive, low-variance features using a dual-loss CNN architecture, improving anomaly and novelty detection performance.

Contribution

It presents a novel deep learning framework with compactness and descriptiveness losses, enhancing feature learning for one-class classification tasks.

Findings

Significant improvements over state-of-the-art in anomaly detection

Effective feature learning with low intra-class variance

Robust performance across multiple datasets

Abstract

We propose a deep learning-based solution for the problem of feature learning in one-class classification. The proposed method operates on top of a Convolutional Neural Network (CNN) of choice and produces descriptive features while maintaining a low intra-class variance in the feature space for the given class. For this purpose two loss functions, compactness loss and descriptiveness loss are proposed along with a parallel CNN architecture. A template matching-based framework is introduced to facilitate the testing process. Extensive experiments on publicly available anomaly detection, novelty detection and mobile active authentication datasets show that the proposed Deep One-Class (DOC) classification method achieves significant improvements over the state-of-the-art.