High Performance and Cost-Effective Video Transcoding for Heterogeneous Networks

• Main Authors: Ching-Ting Hsu, 徐敬亭
• Other Authors: Mei-Juan Chen
• Format: Others
• Language: en_US
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/83015828482584079362

博士 === 國立東華大學 === 電機工程學系 === 97 === Universal multimedia access is emerging for many multimedia services such as videoconferencing, distance learning and video on demand. These applications require a variety of devices, access links and resources. One of the key technologies is video transcoding that converts a pre-coded video bitstream into different formats to satisfy the user’s requirements. The format of a transcoded video could be defined as bitrate, frame rate, spatial resolution, coding syntax and content. In this dissertation, we aim at the performance and efficiency enhancements of the video transcoder over heterogeneous networks. First, the frame rate optimization scheme for temporal-domain video transcoders is discussed. We use Lagrange optimization method to trade-off frame complexity, channel bandwidth and frame rate for determination of the optimized frame rate. Since the straightforward cascaded pixel domain video transcoder is too complicated to implement, the motion composition schemes for temporal and spatial video transcoders are respectively presented in the second part of this dissertation to reduce the computational complexity of the video transcoder. The Discrete Cosine Transform (DCT) is utilized to measure the significance of the pre-coded motion vectors. On the other hand, for the heterogeneous networking environments, the error resilience tools are utilized to protect the video bitstream. Hence, the error resilience video transcoding technique which analyzes the channel condition and video content is proposed in the third part of this dissertation to adjust the intra macroblock in a frame. Finally, an effective H.264 to MPEG-4 video transcoder is proposed. Experimental results show that the proposed algorithms can effectively enhance the performance of the video transcoder. Thus, the video media universal access can be realized.