Chapter 13
Features of OVID

---

• Video Objects as Central Units
  
  
  • The oid of the video object
  • a set of pairs of starting video frame numbers and an ending video frame numbers of the video scene
  • a set of attribute/value pairs which describe the contents of the corresponding video frame sequence
  • basic operations such as play, inspect,and disaggregate
  
  
• Dynamic and Incremental Object Identification and Definition
  
  
  • Since the video object data model does not assume the existance of a database schema, users should be able to identify a meaningful scence at any time and define the meaningful scene with its descriptional datas a video-object. Thus, the system should allow users to add necessary attributes as well as to use attributes of predefined video-objects. Currently, the attributes and attribute values are given to each object interactively in a manual fashion.
  
  
• Video Object Composition
  
  
  • Facilitate the definition of a video objects which corresponds to more than one consecutive sequence of video frames. This operation corresponds to merge operation descibed earlier.
  • Inheritance of descriptional data is implemented in this model.
  • Supports the creation of a video-object by applying the overlap operation to predefined video-object. Some descriptional data for the predefined video-object are also inherited by the overlapped video-object.
  
  
• Browsing by Video Object Disaggregation
  
  
  • To browse a large video-object, OVID generates a decomposition of a video-object into smaller video-objects. This corresponds to the interval projection operation. In OVID, the operation is called disaggregationi, and the result of the disaggregation operation is a collection of smaller video-objects, each of which inherits some descriptional data from the original video-object. The diaggregation operation can be repeatedly applied. This is useful in finding a scene in a large video-object in a navigational manner.
  
  
• Generalization Hierarchy for Atomic Values
  
  
  • OVID supports the use of a generalization (is-a) hierarchy which consists of atomic values that are used as attribute vakues of video-objects in both the phases of video-object creation and retrieval of video-objects.
  
  
• Video Objects as Attribute Values
• Ad Hoc Query Facility for Video Objects
  
  
  • videoSQL: a query facility allows for the retrieval of a collection of video-objects that satisfy a given condition.
  • Support inheritance based on the interval inclusion relationships
  • Support queries to retrieve video-objects that contain a specified vide-object as their attribute value.
  • Spatial and temporal compositions of audio and video objects can be treated.