Abstract
The current level of request for applications involving content-based video coding and video information retrieval is increasing the relevance of systems able to somehow ‘understand’ the content of video sequences. This type of systems will enable the provision of services where enhanced user interaction with the visual content is supported. Such an object-based video coder is being specified by ISO under the MPEG-4 project [1], while the standardization of video description capabilities to support content-based video indexing and retrieval is being considered by MPEG-7 [2]. In this paper, we discuss the issue of video content identification and characterization, and the importance of providing the means for the user to interact with the analysis process so that the achieved results can have a meaningful and powerful semantic value.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have