length whatsoever, as long as it refers to a fixed length of time for any given video. Different
videos in the library
There is now growing interest in organizing and querying large bodies of video data. In this paper, we will develop a may use the same abstract notion of frame to denote differsimple SQL-like video language which can be used not only to ent lengths of time.
identify videos in the library that are of interest to the user, Even when the notion of a frame is fixed, video querying but which can also be used to extract, from such a video in a leads to complications-for example, a user may (inforvideo library, the relevant segments of the video that satisfy mally) ask queries such as: Find me a segment of the specified query condition. We investigate various types of video where a party occurs. This informally user requests and show how they are expressed using our query phrased query admits a number of possible interpretations:
language. We also develop polynomial-time algorithms to process such queries. Furthermore, we show how video presenta-β’ For instance, a party occurs in frames 33 through 56 tions may be synthesized in response to a user query. We show of the movie, The Rope. Should this entire segment of how a standard relational database system can be extended in video be included in the solution or is it enough to show order to handle queries such as those expressed in our language. the user any subsegment?
Based on these principles, we have built a prototype video
β’ Surely, the user should be able to specify that he would retrieval system called VIQS. We describe the design and implelike to see at most say 10 frames of video that satisfy the mentation of VIQS and show some sample interactions with specified query. For example, the user may not wish to see VIQS.