Object Authoring Tool for Video Annotation
In hyperlinked video, objects are selectable resulting in an associated action, akin to linked rich media content about the objects of interests. Possible venues for hyperlinked video include broadcast TV, streaming video and published media such as DVD. Hyperlinked video offers new interaction possibilities with streaming media.
Localization of objects of interests is critical for interactive services in IPTV systems, so that a regular TV broadcast (MPEG-2/-4) is augmented with additional information (MPEG-7 encoded) which defines those objects in the video, along with additional content to be displayed when they are selected. Specification of objects with additional content (metadata) is usually implemented by an authoring tool (video annotation), consisting of such functions as extraction of shots and key frames, specification of the interactive regions, and tracking of the specified regions to get the region locations in all frames.
This annotation tool works mainly at the object level, however open to additional scene level and event level. Users label the bounding box of an object initially and then activiate the tracking module. The tracker can be stoped any time and the tracking result can be edited by mouse control. In case the scene changes, the object is completely occluded for a period of frames, or the tracker fails totally due to drifting errors, the user can reinitialize the tracker again. Eventual results are saved in a meta data file (XML) for use.
Figure 1: Annotation tool initialization.
Figure 2: Video input in Annotation tool. 
Figure 3: Object labeling for tracking initialization. 
Figure 4:Selection of tracked objects for Editing. 
Figure 5: Replay of all edited objects in Annotation tool. Demo Video Video Annotation (0.8M)