Content Representation
The vast majority of today's natural content is captured using traditional cameras and microphones.
The acquisition process fails to capture a significant part of the inherent structure of visual and aural information.
For a long time, the emphasis in representation has been on compression.
Current needs, such as content-based retrieval and browsing, imply object-oriented design approaches.
Current state-of-the-art techniques cannot recognize high-level objects , but low-level salient image regions or audio segments.