MPEG-2 to H.264 TRANSCODING: Why AND how?
Dec 1, 2006 12:00 PM, BY SANTHANA KRISHNAMACHARI AND KYEONG HO YANG
The MPEG-2 compression standard has been widely deployed in video distribution infrastructures, such as cable and satellite networks, as well as in several consumer applications, such as DVDs and DVRs. For more than 10 years, end-to-end systems have existed, with several million interoperable encoders, multiplexers and set-top boxes deployed. The need to transcode to and from the MPEG-2 format has arisen.
The need for transcoding
The following three scenarios describe the need for transcoding. First, improvements in compression techniques have resulted in mature new standards that offer significant bit rate gains of 30 percent to 50 percent over MPEG-2. Investments made in legacy devices that can only handle the MPEG-2 format and the existence of large amounts of content that was created using MPEG-2 necessitate that multiple standards will coexist for several years before the eventual conversion to a single standard.
Figure 1. Format transcoding for IPTV architectures
Click image to enlarge.
Second, emerging IPTV deployments of video over bandwidth constrained last mile will result in part of the content distribution chain migrating to H.264, thereby creating a need for efficient transcoders. (See Figure 1.)
Third, new applications, such as high-definition video and real-time broadcast video over handheld mobile devices, require that the same content be made available in several spatial resolutions and frame rate formats. This can be achieved by recreating the same content in several formats. Alternatively, it will be more efficient to create the content once and transcode it to different formats and resolutions as needed. (See Figure 2.)
Finally, improvements in programmable processor technologies, such as the DSPs and FPGAs, have made it possible for video processing vendors to field products that can handle multiple existing formats and field-upgrade the products for future emerging standards.
Types of transcoding
Several forms of transcoding are possible, depending on the specific parameters of the compressed bit stream that are modified during the transcoding process. They include:
Figure 2. Resolution transcoding to support multiple display devices
- Bit rate transcoding
This process changes the bit rate of the compressed bit stream while keeping the resolution, frame rate and the encoding format the same. MPEG-2 bit rate transcoders, also called rateshapers, have been widely deployed today, and they employ efficient, high-density rateshaping by primarily operating in the discrete cosine transform (DCT) domain.
- Format transcoding
This entails converting the compression format — for example, converting an MPEG-2 bit stream to an H.264 bit stream.
- Resolution transcoding
This involves the conversion of coded spatial resolution — for example, converting a standard-resolution bit stream to common intermediate format (CIF) resolution for a mobile application.
| Want to use this article? Click here for options! |  |
