Critical Review for One-class Classification: recent advances and the reality behind them

TL;DR

This paper critically reviews one-class classification (OCC), highlighting recent advances, methodologies across data types, and exposing flaws in current state-of-the-art image anomaly detection algorithms that do not align with OCC principles.

Contribution

It provides a systematic synthesis of OCC strategies, critiques current SOTA image anomaly detection methods, and clarifies the distinction between OCC and other classification approaches.

Findings

SOTA image AD algorithms often do not adhere to OCC principles.

Top algorithms on CIFAR10 are not true OCC methods.

Binary/multi-class classifiers should not be compared with OCC.

Abstract

This paper offers a comprehensive review of one-class classification (OCC), examining the technologies and methodologies employed in its implementation. It delves into various approaches utilized for OCC across diverse data types, such as feature data, image, video, time series, and others. Through a systematic review, this paper synthesizes promi-nent strategies used in OCC from its inception to its current advance-ments, with a particular emphasis on the promising application. Moreo-ver, the article criticizes the state-of-the-art (SOTA) image anomaly de-tection (AD) algorithms dominating one-class experiments. These algo-rithms include outlier exposure (binary classification) and pretrained model (multi-class classification), conflicting with the fundamental con-cept of learning from one class. Our investigation reveals that the top nine algorithms for one-class CIFAR10 benchmark are not OCC. We ar-gue that binary/multi-class classification algorithms should not be com-pared with OCC.