The amount of contents we can obtain through TV broadcasting by satellite, CATV, and the Internet is increasing day by day. In this paper, we propose a method and an application for future Interactive TV service, called ADTV. In our method, users not only get information about specific video objects, but they can describe video objects about them using the incomplete description of portion in each frame. We discuss automatic structuring of video objects automatically using incomplete description. We implement the prototype system to provide users with some basic functions such as "Description" , "Question and Answer", and "Retrieval" regarding real-world video based on 4igital maps. The system can link video objects to digital maps so that users can retrieve an image of a building or get information about an unknown building in a video.
|Number of pages||9|
|Journal||Kyokai Joho Imeji Zasshi/Journal of the Institute of Image Information and Television Engineers|
|Publication status||Published - 1999|
All Science Journal Classification (ASJC) codes
- Media Technology
- Computer Science Applications
- Electrical and Electronic Engineering