A neural network based transcoder for MPEG2 video compression
H.C. Fu, Z.H. Chen, Yeong-Yuh Xu, C.H. Wang · 1999
We propose a neural network method for the high efficiency requatization in the design of a transcoder. In our design, there are two types of video bit rate control in the proposed transcoder. One is the global adjustment of the quantizer scales in which the adjustment is based on the complication of the whole frame, the other is the adaptive adjustment of the quantizer scales, where the adjustment results in a complication of the current macroblock. From the experimental results, the prototype transcoder can achieve a desirable bit rate (1.5 Mbps) with an acceptable image quality. In addition, we constructed a video multiplexer for pay per view (PPV) or near video on demand (NVOD) applications on the proposed transcoder.