| Summary: | 碩士 === 國立東華大學 === 電機工程研究所 === 87 === Anticipating the rapid convergence of telecommunication, computer, and TV/film industries, the just finalized MPEG-4 standard from ISO standardizes algorithms and tools for coding and flexible representation of audio-visual data to meet the challenges of future multimedia applications. In addition to standard MPEG-1 or MPEG-2 like provisions for efficient coding of conventional image sequence, MPEG-4 enables an efficient coded representation of the video data that can be content based, with the aim to use and present the data in a highly flexible way. To enable the envisioned content-based interactive functionalities, the MPEG-4 video standard introduces the concept of video object planes (VOPs). It is assumed that each frame of an input video sequence is segmented into a number of arbitrarily shaped image regions- each of the regions may possibly cover particular image or video content of interest, i.e., describing physical objects or content within scenes. In contrast to the video source format used for the MPEG-1 and MPEG-2 standards, the video input to be coded by the MPEG-4 is thus no longer considered a rectangular region. The input to be coded can be a VOP image region of arbitrary shape and the shape and location of the region can vary from frame to frame. The concept of VOPs involves the problem of shape representation, motion estimation and texture coding. The purpose of using shape is to achieve better subjective picture quality, increased coding efficiency as well as an object-based video representation. There are mainly two categories for binary shape coding: bitmap-based and contour-based.
In this thesis, we propose the quadtree-based shape coding, which exploits the spatial redundancy for bitmap-based shape coding category. The hierarchical structure provides good performance. For contour-based category, the new scheme with polygon approximation in polar coordinates is conducted. The vertex selection and area distortion criterion techniques are included. Computer simulation is conducted. The proposed shape coding algorithms are expected to make compression more efficient and degradation due to errors more graceful. The proposed schemes provide good rate-distortion trading-off. The results indicate that the subjective quality is improved compared to other shape coding methods
|
|---|
