Video Surveillance Quality Assessment Dataset (VSQuAD)

We introduced a new comprehensive Video Surveillance Quality Assessment Dataset Release-2 (VSQuAD-2) dedicated to Video Surveillance (VS) systems. The database consists of original videos recorded by the team members and other sequences selected from public databases. In contrast to other similar datasets, this one contains much more videos with additional distortions and diversified contents with more common scenarios of surveillance scenes. The videos have been artificially degraded with various types of distortions (single distortion or simultaneous multiple distortion) at different levels. In order to improve the efficacy of the surveillance task and the versatility of the existing video quality assessment datasets, night vision CCTV videos are also included. Furthermore, a comprehensive analysis of the content in terms of diversity and challenging problems is presented in this study. The interest of such a database is twofold. First, it serves as a benchmark for comparing different distortion detection and classification algorithms in video sequences. The second interest is to serve as data for the development of learning models for various challenging VS problems such as identification and removal of the most common distortions. The complete dataset is made publicly available with all the technical details when the paper is accepted.