Methods and needs for transcoding MPEG-4 fine granularity scalability video
Yongqing Liang, Yap‐Peng Tan · 2003
We discuss the methods and needs for transcoding MPEG-4 fine granularity scalability (FGS) video. FGS is a scalable video-coding technique recently adopted by the MPEG-4 compression standard. The main feature of FGS coding is that a part of the video bit stream, namely the enhancement layer, can be truncated at any desirable number of bits to support continuous quality/bit-rate scalability. However, not every decoder would support FGS functions and the coding efficiency of FGS video is known to be inferior to that of non-scalable video. This paper proposes methods for transcoding FGS video and restoring, as much as possible, the coding efficiency that is lost in FGS coding. In particular, it is shown that there are cases where extra gain of video quality is possible by exploiting the FGS enhancement layer in the transcoding process. To some extend, the proposed transcoding methods can support spatial scalability that is lacking in FGS coding. Experiments have been conducted and the results are reported to demonstrate the advantages of the proposed transcoding schemes for FGS video.