PseudoBound: Limiting the anomaly reconstruction capability of one-class classifiers using pseudo anomalies
This addresses a specific bottleneck in video anomaly detection for surveillance and security applications, offering an incremental improvement over existing reconstruction-based methods.
The paper tackles the problem of autoencoders in one-class classification for video anomaly detection reconstructing anomalous data too well, which reduces performance, by proposing a training mechanism that incorporates pseudo anomalies to limit this capability, achieving state-of-the-art or competitive results on three benchmark datasets.
Due to the rarity of anomalous events, video anomaly detection is typically approached as one-class classification (OCC) problem. Typically in OCC, an autoencoder (AE) is trained to reconstruct the normal only training data with the expectation that, in test time, it can poorly reconstruct the anomalous data. However, previous studies have shown that, even trained with only normal data, AEs can often reconstruct anomalous data as well, resulting in a decreased performance. To mitigate this problem, we propose to limit the anomaly reconstruction capability of AEs by incorporating pseudo anomalies during the training of an AE. Extensive experiments using five types of pseudo anomalies show the robustness of our training mechanism towards any kind of pseudo anomaly. Moreover, we demonstrate the effectiveness of our proposed pseudo anomaly based training approach against several existing state-ofthe-art (SOTA) methods on three benchmark video anomaly datasets, outperforming all the other reconstruction-based approaches in two datasets and showing the second best performance in the other dataset.