Statistical Techniques for Video Analysis and Searching

With the growing amounts of digital video data, effective methods for video indexing are becoming increasingly important. In this chapter, we investigate statistical techniques for video analysis and searching. In particular, we examine a novel method for multimedia semantic indexing using model vectors. Model vectors provide a semantic signature for multimedia documents by capturing the detection of concepts broadly across a lexicon using a set of independent binary classifiers. We also examine a new method for querying video databases using interactive search fusion in which the user interactively builds a query by interactively choosing target modalities and descriptors and by selecting from various combining and score aggregation functions to fuse results of individual searches.