| 摘要 |
Subtitling can be based on either pixel data or on character data. Character data allow very efficient encoding, but from character strings alone, subtitling can not be converted into a graphical representation to be overlaid over video. The intended character set, font and e.g. font size, must either be coded explicitly within the subtitling bitstream or an implicit assumption must be made about them. In pixel-based subtitling, subtitling frames are conveyed directly in the form of graphical representations by describing them as (typically rectangular) regions of pixel values on the AV screen, at the cost of considerably increased bandwidth for the subtitling data. According to the invention, a font memory is used that allows an efficient realisation of pixel-based subtitle lettering, because the glyphs need only be transmitted once and thereafter are referenced by relatively compact character references during the AV event. Thereby the invention combines the advantages of pure pixel-based and pure-character-based subtitling schemes, while mostly avoiding their respective shortcomings. <IMAGE> |
