Unmasking the abnormal events in video
This addresses the problem of detecting anomalies in video for surveillance or monitoring applications, offering an unsupervised approach that is incremental in adapting an existing method to a new domain.
The authors tackled abnormal event detection in video without needing training sequences by adapting the unmasking technique from text authorship verification, achieving state-of-the-art results on four benchmark datasets and running at 20 frames per second.
We propose a novel framework for abnormal event detection in video that requires no training sequences. Our framework is based on unmasking, a technique previously used for authorship verification in text documents, which we adapt to our task. We iteratively train a binary classifier to distinguish between two consecutive video sequences while removing at each step the most discriminant features. Higher training accuracy rates of the intermediately obtained classifiers represent abnormal events. To the best of our knowledge, this is the first work to apply unmasking for a computer vision task. We compare our method with several state-of-the-art supervised and unsupervised methods on four benchmark data sets. The empirical results indicate that our abnormal event detection framework can achieve state-of-the-art results, while running in real-time at 20 frames per second.