HD3D: It's Coming Fast

Spurred on by the technical and economic success of digitally produced and distributed motion pictures, the inevitable emergence of the third dimension of consumer HD is upon us. Consider the following items:

Approximately 5000 movie screens in North America now show motion pictures digitally distributed with approximately 1000 of these 3D capable projection systems. However, this estimate is gradually being increased, as the digital production and distribution cost model becomes increasingly attractive.

At least ten major motion pictures, including all of DreamWorks' animated productions will be released in 3D within the next two years. Many in the industry feel this number is probably low due to the increasing availability of economically attractive digital 3D production tools.

Many Hollywood production support companies are providing Digital 3D services. Among them are PACE, 3Ality Digital, Dolby, RealD, Thomson and others. Adding 3D to an existing digital production physical infrastructure is relatively inexpensive. Some estimates are in the 10% to 15% range. However, as experience with 3D is gained, specific camera and editing skills are being developed that greatly enhance the 3D viewing experience and move it away from the early film 3D "gimmick" exhibitions. These techniques include fewer editing "jump cuts," less aggressive panning and more discreet use of "focused object" (sometimes called "convergence") , thereby "opening up" the scenes to provide more realistic 3D images.

Experiments in the production and distribution of live HD3D are on-going with HD Net and others. In the past 18 months two National Basketball Association games have been produced live in HD3D and transmitted to closed circuit venues. Both CBS and Time Warner's HBO have been experimenting with HD3D as has ESPN with tests of HD3D segments of their "X-Games" production last August. Additionally, many recently digitally produced or transferred 3D material exists such as IMAX 3D features. More recent 3D hits such as Chicken Little, Meet the Robinsons, Journey to the Center of the Earth, Beowulf, Hannah Montana and U2 3D are "in-the-digital can."

Several television commercial production equipment manufactures are starting to offer digital 3D cameras, editing and monitoring equipment. These include Thomson (Glass Valley), Sony, Avid, Quantel, Cine-tal, RED and others.

Panasonic, Samsung, Mitsubishi, Philips, Hyundai and others are currently publically demonstrating or offering HD3D consumer display systems. The CE industry is expected to sell up to two million such 3D capable ("ready") HDTV monitors this year. Most of these are DLP based rear projection models that employ switched frame techniques generating the stereoscopic images at a sequential 48p or 60p frame rate. Source material includes 3D encoded DVD's and games played-back via computer (with a specific video card and special software) and coupled to the display monitor via DVI/HDMI. The viewers wear IR coupled "shutter" glasses to product separate synchronized images for each eye, thus producing the 3D effect. (Philips, however, uses an "autostereoscopic" system which does not require glasses). However, as more 3D movie productions are transferred to HD3D on Blu-ray discs, one can understand the economic potential of HD3D.

Standards: The real key to moving HD3D forward, however, is the generation of acceptable production, transmission and reception/display standards for stereoscopic high definition television. To this end, last summer the Society of Motion Picture and Television Engineers (SMPTE) established a task force to define the parameters for 3D mastering of content for viewing in the home. This is the prologue to actually generating SMPTE consumer HD3D display standards. Further, this fall, the Consumer Electronics Association (CEA) will form a discussion group to investigate the need for standards to address the delivery and presentation of consumer HD3D television. Thus, both the "software and hardware" aspects of the industry are starting to focus on generating the standards that, like radio, television, color television, stereo audio and HDTV will serve as the technical foundation to drive the future of HD3D consumer products. Quite possibly the two efforts will merge under the auspices of the ATSC (Advanced Television System Committee) similar to the process used to generate the US DTV standard. But the gestation period for standards is normally very long, and unless they are expedited in this case, the technical and market pressures may result in another HD DVD - Blu-ray type debacle.

The good news is that the addition of the stereoscopic element will not require much change in the present HDTV production, transmission and reception/display infrastructure. Yes, stereoscopic cameras will be needed along with 3D editing suites, but the added information to deliver 3D video can essentially "pass through" the existing HDTV production and distribution system. For instance, the two stereoscopic channels can be mixed, compressed and added to the baseband video digital stream. The two components needed include one spatial (the distance between the pixels of the two images), and one temporal (the rate of change of the pixel spacing as the image moves). The multiplex composite of these two components can then be compressed and added to the brightness ("Y") baseband digital stream. This would be done at a very low level and only discernable by a dithering of the least significant digit in the decoded "Y" stream. In this manner, similar to a "dynamic" invisible copy protection watermark, the 3D information would (hopefully) survive channel distribution.

Non 3D receivers would ignore the 3D information, but it would be detected by 3D sets, decoded and thus establish the spatial position of the two images. Using this type of scheme would require no changes in the production infrastructure and would be transparent to existing HD accessories such as HD disc players and DVRs.

In terms of HD3D development, the production elements are fairly mature. HD3D cinematography and editing techniques continue to be refined but are certainly within the dimensions of commercial acceptance. Psycho-optic 3D elements that cause viewer discomfort such as eye strain and unrealistic image separation are understood and being addressed.

However, the display side needs work. Although shutter glasses are acceptably, the "Holy Grail" continues to be some type of wide aspect ratio direct view system that would be compatible with the legacy switched (glasses) approach. Perhaps some holographic techniques would work. Nonetheless all the "big boys in the pool room" of the industry are gambling bunches of dollars on it. Next year's CES and NAB exhibitions are sure to see the real start of the big HD3D push. It will only get better.

By Ed Milbourn, HDTV Magazine