Hypermedia Data Modeling, Coding, And Semiotics - Proceedings of the IEEE

than other forms of communication. This is because there is no clear separation of form or expression from the content in music since the expression is the content. Instead of information, it mainly communicates emotion since music is the logical expression of feelings. Music also has a powerful referential potential, which assigns meaning through association to past experiences. Music has Table 6 a highly evolved structure with definite rules much like normal grammars, resembling the hierarchical organization of text. While music can be physically expressed in terms of frequency, time, and intensity, in musicological terms, music has three dimensions: melody, harmony, and rhythm. Melody is the progression of tones produced by adding them horizontally, while adding tones vertically generates chords and adding chords sequentially produces harmony. Rhythm is produced by periodic repetition. The smallest subsign of music is therefore a tone or toneme. While a single tone has no embodied meaning, a short series of tones can readily convey an emotional experience [113]. If the series is ascending, it expresses outgoing emotion; if descending, it expresses incoming emotion. If it is in a major key, joy is conveyed; alternatively, sorrow [114]. The logical theory of semiotics in music [115] postulates that at least three notes are required to form monemes. Articulation also exists in pictures. Various proposals for the articulatory units include the concept of chromemes (color elements) and formemes (shape elements), among others [116], [117]. However, it is difficult to foresee any lexical constructs to combine chromemes to produce meaningful units. Another approach that has been proposed is Marr’s model [118], which postulates the existence of three different representation systems starting with an initial 2-D primal sketch and progressing to a viewercentric 2 1/2-D sketch and finally to an object-centric 3-D representation for semantic recognition. Apart from the vague notion of texturemes, which are difficult to isolate and are not distinctive, Marr’s model does not really provide primitive elements that could be considered suitable articulatory units. A better approach is based upon Gestalt psychology, although a suitable definition of subsigns or primitive elements in pictures is currently lacking. Video or film communication is a composite medium of a sequence of images undergoing motion. Since the articulation within each frame is the same as for still images, the primary feature of video and film is its temporal domain. Accordingly, the first level of articulation is the shot (or scene) and is known as the videme [119], [120]. Some uncertainty has been expressed regarding the existence of a second level of articulation. One proposal is that it is composed of spatial-graphical objects called cinemes (or iconemes). These already represent meaningful elements, however, and are therefore unsuitable. An alternative would be to exploit the predominantly temporal nature of video and the observation that interframe changes are largely motion induced. Accordingly, it would not be inappropriate to propose that the second-level units be composed of a set of motion primitives. Eco [121] considers motion primitives the dynamic units of a third level of articulation called