Automatic Behavior Understanding in Crisis Response Control Rooms

摘要: This paper addresses the problem of automatic behavior understanding in smart environments. Automatic is defined as generation semantic event descriptions from machine perception. Outputs available perception modalities can be fused into a world model with single spatiotemporal reference frame. The then used input by reasoning engine that generates descriptions. We use newly developed annotation tool to generate hypothetical outputs instead. applied based on fuzzy metric temporal logic (FMTL) and situation graph trees (SGTs), promising universally applicable tools for understanding. presented case study report staff training purposes crisis response control rooms. Various group formations interaction patterns are deduced person tracks, object information, information about gestures, body pose, speech activity.