Automatic textual description of interactions between two objects in surveillance videos

Abstract The purpose of our work is to automatically generate textual video description schemas from surveillance scenes compatible with police incidents reports. Our proposed approach based on a generic and flexible context-free ontology. general schema the form [actuator] [action] [over/with] [actuated object] [+ descriptors: distance, speed, etc.]. We focus containing exactly two objects. Through elaborated steps, we formatted description. try identify existence an interaction between objects, including remote which does not involve physical contact point out when aggressivity took place in these cases. use supervised deep learning classify into or no-interaction classes then subclasses. chosen descriptors used represent subclasses are keys systems that help live alerts facilitate offline investigation.