Video analysis and interpretation
The analysis and interpretation of video sequences is a growing research field in the scientific community due to its many practical applications in real contexts.
The MIVIA lab is active in this research field by more than a decade, and has defined innovative techniques for the real time processing of video sequences for different applicative scenarios.
In particular, the interests of the group are focused to the following areas:
- intelligent video surveillance systems: the group has proposed innovative techniques for the automatic detection of events of interests from video taken by surveillance cameras deployed in indoor and outdoor environments, with solutions for the reliable detection and tracking of objects moving within the scene, also in difficult situations characterized by multiple occlusions, shadows and presence of reflecting floorings;
- crowd analysis: the research is focused at estimating the number of persons or more in general of objects of interest which are present in an area framed by a surveillance camera, with the aim of detecting dangerous situations or providing overall statistics about the scene;
- behavior analysis and interpretation: the group has proposed techniques for the automatic analysis of the behavior of the objects moving into the scene taken from a camera with the aim of detecting anomalous situations or patterns of common behaviors.