Content-based retrieval of video data by the grammar of film

Image/video data contains numerous information such as objects, image attributes (e.g., color, height, width) of objects, and various semantic contents. Provision of a scheme for the retrieval of video data content is still one of the unfulfilled issues. The authors propose a method of retrieving video data from films by way of specifying the semantic content of scenes. Related to the creation of films, there is a so-called “the grammar of film”, which is an accumulation of knowledge and rules for expressing certain semantics of a scene more effectively. They concentrate on the features of video that can be observed as a consequence of the effects of the grammar of film. Since the features relate to shot length, the degree of image dynamics of shots, or the combination of similar shots, this method is less time-consuming than other ways of evaluating sequences of images in detail