作者: N. Brady
DOI: 10.1109/76.809154
关键词: Discrete cosine transform 、 Image processing 、 Image texture 、 Motion compensation 、 Arithmetic coding 、 Theoretical computer science 、 Computer science 、 Transform coding 、 Computer vision 、 Artificial intelligence 、 Data compression 、 MPEG-4
摘要: MPEG-4 is the most recent standard for audio-visual representation to be published by International Organization Standardization. One of many new features its ability represent two-dimensional video objects arbitrary shape. For this purpose, uses conventional motion-compensated discrete cosine transform syntax color/texture coding and augments with an explicit compressed object's This paper intended as a tutorial in means encoding decoding arbitrarily shaped specified MPEG-4. The major emphasis on explaining compression technology associated normative shape representation, i.e., block-based context-based arithmetic encoding, but some aspects texture are also highlighted. specifications presented informal way, motivations underlying algorithm clarified. In addition, effective methods suggested performing nonnormative tasks, several performance tradeoffs illustrated.