| Summary: | 碩士 === 國立東華大學 === 資訊工程學系 === 87 === Advances in digital video technology and storage devices have made it possible to use digital video into a number of multimedia applications. In 1994, Moving Picture Expert Group (MPEG) initiated a new standard, referred to as MPEG-4. Unlike the frame-based MPEG-1 and MPEG-2, MPEG-4 emphasizes on coding efficiency and allows for content-based access and transmission of arbitrary shaped object. MPEG-4 standard allows the encoding of video object using shape coding, motion estimation, and texture coding information for interactivity, high compression, and scalability. Among these codings, binary shape coding method includes two major categories: one is bitmap-based shape coding and the other is vertex-based shape coding.
This thesis proposes a new vertex-based shape coding method for encoding object's vertices of a polygonal approximation by adapting the representation to the dynamic range of the relative locations with different approximation distortion. The octant-based representation is used to represent the relation of the adjacent vertices, and obtains the information of relative location with adjacent vertices. Simulation results demonstrate that the new vertex-based shape coding may reduce more bits for closely spaced vertices.
|
|---|
