Reducing False Alarms in Video Surveillance by Deep Feature Statistical Modeling
This work addresses the critical issue of false alarms in video surveillance, which is essential for practical deployment, though it is incremental as it builds on existing change detection methods.
The paper tackled the problem of high false alarm rates in unsupervised video change detection by proposing a method-agnostic a-contrario validation process based on deep feature statistical modeling, which significantly reduced false alarms at both pixel and object levels across multiple datasets and methods.
Detecting relevant changes is a fundamental problem of video surveillance. Because of the high variability of data and the difficulty of properly annotating changes, unsupervised methods dominate the field. Arguably one of the most critical issues to make them practical is to reduce their false alarm rate. In this work, we develop a method-agnostic weakly supervised a-contrario validation process, based on high dimensional statistical modeling of deep features, to reduce the number of false alarms of any change detection algorithm. We also raise the insufficiency of the conventionally used pixel-wise evaluation, as it fails to precisely capture the performance needs of most real applications. For this reason, we complement pixel-wise metrics with object-wise metrics and evaluate the impact of our approach at both pixel and object levels, on six methods and several sequences from different datasets. Experimental results reveal that the proposed a-contrario validation is able to largely reduce the number of false alarms at both pixel and object levels.