Yoshinobu Tonomura

Content oriented visual interface using video icons for visual database systems

This paper clarifies the importance of content oriented visual user interfaces using video icons for visual database systems. Effectiveness of both still and live video images, especially for users browsing and interaction, is shown by usage of the MediaBENCH (hyperMedia Basic ENvironment for Computer and Human interactions) which is a basic prototype multimedia database system. Various methods of handling video data on the MediaBENCH are introduced and discussed to show how video data can be manipulated on visual database systems which deal with spatial and temporal factors. A visual interface using video icons is quite suitable to video editing systems, video mail systems, or other electronic video document systems. Brief application profiles are shown for guidance.

Scene retrieval method using temporal condition changes

This paper proposes a method that retrieves a required scene with such dynamic changes as the motion of the object in the scene as the key from the dynamic images stored as the database. In the traditional database, scenes are retrieved using the keyword for each scene, for example. By contrast, in the proposed system the object of retrieval is an arbitrary interval represented by the dynamic images, and the images are retrieved based on the content represented by the dynamic images. The retrieval request from the user is transformed into the functional format, which describes the time-course of changes of the object in the image. The retrieval is based on the primitive characteristic data (PCD) as the management data for the video images, and the condition description factor (CDF) obtained by the transformation of the retrieval request from the user. n n n nAn experiment is attempted where the two-dimensional position coordinate of the object on the scene is used as PCD, and the motion of the object in the scene or the relative motion of more than one object is used as the key to retrieve the desired scene. The effectiveness of the method is verified.