出版社：The Institute of Image Information and Television Engineers
摘要：A layered video multicast scheme using multiple channels has been getting much attention for Internet/Intranet video broadcast. In this paper, a transmission control method is analyzed for the layered coded video. This method uses temporal layer separation based on the dependency between the I, P, and B prediction modes and on the spatial layer separation by data partitioning of the DCT coefficients. In order to avoid an overflow/underflow in a receiver buffer, a transmission starting time and a stand-by duration for a decoder, before starting decode process, are given as transmission control parameters. The buffer size required for the receiver as well as the delay time due to buffering are also given. The effect of data partitioning of the I-frame on video quality is evaluated and bitrate allocation between layered video streams are considered through computer simulation to achieve an optimum quality.
