The Study of Video Trans-coding in Transform Domain

• Main Authors: Chen-Feng Yang, 楊承峯
• Other Authors: Chung-Ming Kuo
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/07946247705939208366

碩士 === 義守大學 === 資訊工程學系碩士班 === 98 === Recently, there has been a rapid growth of multimedia access devices (e.g., computer, cellular, PDA and television systems) due to the development of video compression and networking technologies. Therefore, it is an important issue that converts the compressed information between delivery systems and the terminal devices in the heterogeneous networks. Therefore, various video trans-coding techniques are developed. In this thesis, we focus on exploring frame downsizing of video trans-coding and motion vectors re-estimation in DCT domain. In frame downsizing, the odd and asymmetrical rate is proposed to solve the problem of traditional scheme which is available only for even rate downscaling. In order to decrease the computational complexity and improve the accuracy, in our works, the downscaling and motion vector re-estimation are both perform in DCT domain. Simulation results show that the proposed method can efficiently achieve better performance for both visual quality and PSNR.