Spot The Odd One Out: Regularized Complete Cycle Consistent Anomaly Detector GAN

TL;DR

This paper introduces RCALAD, a novel GAN-based anomaly detection method that uses cycle consistency and a new discriminator to improve accuracy and training efficiency across diverse datasets.

Contribution

The paper proposes RCALAD, a new anomaly detection framework with a specialized discriminator and input space distribution, enhancing detection accuracy and training stability.

Findings

Outperforms existing models on six datasets

Achieves more consistent class-wise accuracy

Demonstrates improved anomaly separation

Abstract

This study presents an adversarial method for anomaly detection in real-world applications, leveraging the power of generative adversarial neural networks (GANs) through cycle consistency in reconstruction error. Previous methods suffer from the high variance between class-wise accuracy which leads to not being applicable for all types of anomalies. The proposed method named RCALAD tries to solve this problem by introducing a novel discriminator to the structure, which results in a more efficient training process. Additionally, RCALAD employs a supplementary distribution in the input space to steer reconstructions toward the normal data distribution, effectively separating anomalous samples from their reconstructions and facilitating more accurate anomaly detection. To further enhance the performance of the model, two novel anomaly scores are introduced. The proposed model has been thoroughly evaluated through extensive experiments on six various datasets, yielding results that demonstrate its superiority over existing state-of-the-art models. The code is readily available to the research community at https://github.com/zahraDehghanian97/RCALAD.