The consumer-electronics industry is pushing 3D faster than any previous technology, according to Rick Dean, chair of the 3D@Home Consortium, at the recent 3D@Home Workshop in San Jose, Calif. “Considering the fact that we are only two years into this, it is phenomenal that we already have something getting into the home and there is a content workflow coming and a commitment to make it happen on both the theatrical side and home side,” he added.
But many details still need to be worked out to create a consistent 3D infrastructure. One of the biggest questions facing the industry is how to provide enough underlying consistency so that it works seamlessly across broadcasters, cable operators, set-top boxes and TVs. Chris Chinook, founder and president of Insight Media, said there are challenging technical problems in getting equipment from different vendors to work together.
For example, there are multiple options for compressing and packaging 3D content. A 3D video requires more data than does a traditional 2D video, yet operators want to leverage the traditional video-distribution infrastructure. Both Blu-ray disc players and Internet distribution can support higher data rates relatively easily, but cable providers and terrestrial broadcasters are wondering how to squeeze the additional bits into existing HD channels.
To help address these challenges, vendors have developed a variety of 2D-video, frame-compatible approaches. Each approach uses an algorithm to extract specific parts of the image and repack this data into a traditional video image format. The three main approaches being explored today filter out alternate rows, columns or checkerboard–like blocks to reduce the size of the image.
Once reduced, these two images are packaged together. The most common configurations are to pack one image above the other or to arrange them side by side, because these dual images can be compressed more efficiently using MPEG than other patterns. These same packing arrangements are also applied to images extracted via a checkerboard pattern, which can be a problem for the 3D decoder because the same side-by-side arrangement is used for checkerboard or column filtered video. Chinook explained, “If the device sees two images side by side, there is no way to know how it was packed.”
Vendors are starting to incorporate a variety of types of metadata into 3D data formats, which could help set-tops to identify the type of stream automatically and to fine-tune the content for the specific TV. This metadata also could be used for adjusting the subtitles to the appropriate depth in the current video. But there are at least three different metadata approaches, so a device vendor would have to license the technology for all three in order to support seamless interoperability today, Chinook added.
Although more work needs to be done on the specifications, broadcasters have committed to making a solid step into 3D. ESPN rolled out a 3D sports channel on the Comcast, ATT, Verizon and DirecTV networks, and it plans to show 85 3D events in its first year. DirecTV also has launched a 3D channel, and the Discovery channel is working with IMAX and Sony to roll out a 3D channel in 2011.
Most of these are trial based activities. “Everything is not best-suited for 3D, and we are going to find that out,” Chinook concluded. “We are going to go through an experimental phase, trying out a lot of content, until we figure out what works.”
Foe more on the 3D@Home Consortium, go to www.3dathome.org.