DESCRIPTION: 

Video annotation tool (VAT) is a tool to manually annotate MPEG video files. It provides a user friendly interface for the accurate and undemanding live and "frame by frame" annotation of video. VAT imports its descriptors from predefined OWL ontology files. In addition, VAT enables the definition and employment of user defined descriptors. Three different workspaces opening as different tabs allow for the annotation of the video as a whole, the annotation of specific shots and the annotation of specific regions. As regards to regions, size adjustable region marking boxes enable the captivation of movement trajectories, simply by using mouse dragging and dropping. Originally developed within the BOEMIE research project, VAT has already been successfully used for the annotation of sports events videos. Its augmentation is considered as an ongoing research effort for MKL.