Standardizing 3D Imaging: From Movies to TV Sets
The three-dimensional (3D) content business is growing very quickly indeed, as more and more theaters announce support for 3D imagery. Some films in the US have grossed Yen30 billion. Standardization efforts have started up, primarily in the US, to accelerate this trend, and in 2010 audio-visual (AV) equipment such as televisions will also begin to provide full-scale support for 3D.
"A diverse range of businesses are getting involved in 3D imagery. Everything from content like movies and games to hardware, like TVs. It feels like it's turning into the same sort of thing as the automotive industry," says Mohammad R Ahmadi, president, Technologies & Corporate Development, XpanD Inc of the US, which is involved in upgrading movie theaters to support stereoscopic 3D. He is convinced that the 3D movie market, surging fast, will not fizzle out like it did the last time. Driven by 3D imagery, products across the board are gearing up for support, including TVs, personal video recorders (PVR), broadcast services, mobile phones, personal computers (PC), and game systems.
3D Imagery Infrastructure
There is another difference from the last time: standardization efforts are under way at every place the stereoscopic 3D specification is handled, from video content creation to distribution, AV equipment, viewing glasses and more.
The Society of Motion Picture & Television Engineers (SMPTE) of the US, which is spearheading standardization, is working on not only the Home Master Format specification for the basic 3D imagery format, but also moving the entire 3D imagery industry forward. The Consumer Electronics Association (CEA) of the US, meanwhile, has begun work on standardization of the home AV equipment interface, along with a few other things. The new High-Definition Multimedia Interface (HDMI) 1.4 specification for digital consumer electronics already has the 3D spec built-in.
Experimental 3D broadcasting has already started in numerous places around the world, via cable, satellite broadcasting, etc, and standardization is taking these trials into consideration.
The US is not the only nation involved. In Japan, the Association of Radio Industries & Businesses (ARIB) launched a study group for 3D TV in Sept 2008, which cooperates with organizations such as the SMPTE and CEA. Panasonic Corp and Sony Corp, both of Japan, and other firms are guiding standardization relating to Blu-ray Disc technology. The Camera & Imaging Products Association (CIPA) has already completed standards for 3D photographic data, for example.
Until now, individual manufacturers released equipment to their own specs, and there was essentially no way to distribute 3D video content. The various standardization efforts under way now, however, will have the impact of a superhighway on distribution; it will make it much easier for not only the majors to get into the 3D imagery business, but even small- and medium-scale enterprises, and will lay the groundwork for phenomenal market growth.
In the US, there are several 3D films which have grossed US$200 million to US$300 million each. This is top-class even when compared to a regular film. Yano Research Institute Ltd of Japan predicts that the Japanese domestic market for 3D displays will reach about Yen432 billion in 2019, but this number does not include revenues from 3D content itself. Market scale could be quite a bit larger if content is included.
Reliance on 3D Glasses
In spite of all the hullabaloo, there is one thing to keep in mind: almost all the systems expected to utilize 3D content use 3D glasses to make it possible. The RealD method, Dolby method and XpanD method all use different technologies, but all use glasses. The 3D TVs scheduled to be marketed in 2010 by consumer electronics manufacturers for the home are also expected, for the most part, to rely on 3D glasses.
There are many technologies to achieve 3D imagery, and research into technologies that do not require glasses is also active. Even so, the vast majority of businesses that are making solid progress rely on glasses. The reasons are that glasses make it relatively simple to provide a high-resolution image even from a low-cost display panel originally designed for two-dimensional (2D) imagery, and the viewing field providing a beautiful 3D image is very wide. They fulfill the three crucial conditions for 3D viewing: low cost, high resolution and wide viewing angle.
Differences in 3D Glasses
The problem is that even though the idea of using 3D glasses is the same, there are three general ways to do it, namely the RealD method used in movie theaters, the XpanD method, and the Dolby method. Understanding their differences will clarify the future of 3D television. It is unlikely that the differences in glasses will affect the way audiences watch films in theaters, but when it comes to 3D TV, the technology in the viewing glasses is likely to affect cost, image appearance, etc, depending on how they interact with the display technology.
Consider the XpanD method, more properly known as the active shutter method. Also called the time-division method, the images for the left and right eyes are displayed alternately on the movie screen, display, etc. The viewer dons eyeglasses equipped with special lenses, which have liquid crystal lenses with "shutter" functions to turn transparency on and off. The left and right shutters are activated in time to the control signal from the TV, so that the left eye sees only the image intended for the left eye, and likewise for the right.
The RealD and Dolby methods, on the other hand, are passive, with no electrical drive functions in the glasses. The difference between the two is the method in which the left- and right-eye images are multiplexed. RealD's approach is polarized, with left- and right-eye images divided by right-hand and left-hand polarization and displayed together. Circular polarization filters in the eyeglass lenses make it possible to see the individual images.
Dolby Laboratories Inc of the US instead divides the imagery by color, splitting the visible spectrum into six wavebands, and using two colors each to represent red (R), green (G) and blue (B). The viewing glasses have different RGB filters in the left and right lenses. The anaglyph type of red/blue eyeglasses used decades ago is a primitive form of the same thing. The Dolby technology, however, is currently only used in movie theaters, with no sign of home use at all.
Panel Drive Technology
Manufacturers of home-use TVs, projectors, etc, seem likely to choose either time- or polarization-multiplexed methods for the time being. If the technique is the same, it would at least be possible to use the same type of 3D glasses in both the movie theater and the home. Unfortunately, that's not always the case.
As mentioned above, the 3D glasses used have a significant effect on panel manufacturing cost, resolution and other points. For example, the time-division technique can utilize standard 2D panels, dramatically reducing the cost of implementing 3D support. Depending on the specific drive integrated circuits (IC) chosen, though, there may be resolution issues. 3D rear projection systems sold in North America by companies including Mitsubishi Electric Corp of Japan and Samsung Electronics Co Ltd of Korea, for example, use digital light processing (DLP) from Texas Instruments Inc of the US specifically for rear-projection systems as the display device. DLP is designed to alternate displays of the left- and right-eye images in a checkerboard pattern, which means that for a given frame frequency the image resolution will be half the true resolution of the display device.
The same basic problem occurs with the standard liquid crystal display (LCD) panels used: for a given frame rate, resolution is halved. This is because the drive ICs used for these LCD panels alternate the left- and right-eye images for each horizontal scan line. True, if a dedicated drive IC were developed allowing all the pixels of a single frame to be seen, there would be no resolution loss even without having to boost the frame frequency. Called "frame sequential" technology, it is already used in plasma display panel (PDP) TVs developed for 3D by Panasonic and others.
LCD Panel Response
Panel cost is a key issue with polarization-multiplexed technology. The "micropolarization film" with left and right circular polarization filters in scan line or pixel units demands extremely high-precision alignment when glued to the panel, which is expensive. There have been reports that the vertical viewing angle is reduced when circular polarization filters are positioned for each horizontal scan line.
Even so, the technique is used in products from Hyundai IT Corp of Korea, Sony prototypes, etc, because it can utilize the same inexpensive, battery-free passive 3D glasses as used in RealD. One of the reasons is that time-division just doesn't work smoothly with LCD panels. Flicker is a problem, as is excessive crosstalk, which allows the image meant for one eye to also be seen by the other.
Both of these problems stem from the slow response of standard liquid crystal: specifically, flicker occurs when the 3D glasses open/close shutter speed is so slow that it becomes visible. The same problem occurs in PDP and DLP, both of which have slow shutter open/close operation, but LCD panels suffer from the additional problem that it is more difficult to boost the frame frequency (shutter speed). LCD panels use backlights, which means that the per-pixel emission time is longer than in self-emitting designs such as PDP or DLP, resulting in increased crosstalk.
Improving LCDs
LCD panel manufacturers involved in 3D displays are developing their own solutions to these problems. Samsung Electronics, for example, uses a high (4x) frame rate, in other words 240Hz, and inserts a black image immediately after the left- and right-eye images to shut off emitted light. Flicker is reduced because the higher frame rate makes the speed of change between dark and light too rapid for the human eye to notice. Crosstalk is reduced because the higher frame rate and black image insertion reduce the effective pixel emission time.
Toshiba Mobile Display Co Ltd (TMD) of Japan is developing its own 3D display and glasses, using an optically compensated birefringence (OCB) LCD. OCB LCDs are known to have a very short moving picture response time (MPRT), an index of response, of only 3ms. In addition, says a researcher working for the firm, the insertion of black images between left- and right-eye views and switching off the backlight made it possible to slash the value of the crosstalk ratio to "only 0.08%." Flicker is said to be low even at a 120Hz frame rate. Until now, TMD has only used OCB LCDs in small panels for mobile phones and the like, but recently prototyped a 3D display using medium-size 32-inch OCB LCD panels. No commercialization date has been announced.
Other manufacturers are attempting to resolve the problems using off-the-shelf LCD displays. Olympus Visual Communications Corp of Japan has used a display manufactured by BenQ Corp of Taiwan, with twisted nematic (TN) liquid crystal, and reduced crosstalk with proprietary LCD shutter-type glasses. TN LCD has a much shorter MPRT than standard liquid crystal, about 2ms to 6ms, similar to OCB LCD. In addition, says a representative for the firm, both left and right lens LCD shutters are closed simultaneously in each cycle, providing the same effect as inserting a black image on the panel side.
There are also manufacturers developing new technologies beyond the bounds of existing frameworks, such as LG Display Co Ltd of Korea, which plans to ship a proprietary panel product in 2010 combining polarization- and time-division concepts. An active polarization sheet is mounted in the panel to control circular polarization in pixel units, instead of affixing micropolarization film to the panel surface. This sheet alternately displays left- and right-eye images (time division). It can be used with passive glasses, and according to a researcher for the firm, "The high display efficiency means it is two times brighter than conventional LCD shutter designs."
Towards Standardization
While TV manufacturers are selecting or developing their own 3D technologies, standardization under way at the SMPTE, CEA, etc, may have a major effect on the direction of 3D evolution.
The SMPTE, for example, plans to complete standardization of the basic data format before the end of 2010. A detailed report on the existing data format was released in April 2009, and a dedicated 3D lab was constructed within the Entertainment Technology Center (ETC) of the University of South California (USC) with the goal of (according to the SMPTE) "reviewing the possible utilizations of 3D imagery and creating a world map of the 3D technology." Evaluations are under way of individual technologies, and optimal combinations of technology for various types of content. While the group's results have no binding power over manufacturers, they could well become selection criteria for users.
In contrast, the CEA (handling standardization of the interface for home AV equipment, including TVs) created an R4WG16 working group called to standardize 3D glasses interface, equipment, terminology, etc, on May 26, 2009. According to Brian Markwalter, vice president of Technology & Standard at CEA, "The balance between standardization and wide freedom of technology selection is crucial. When we consider ease of use by viewers it is clear that we can't ignore 3D glasses technology, and we will have to standardize the interface to some extent." Unlike the SMPTE, however, the CEA has not yet set concrete deadlines. While they have no plans to rush ahead, the group's Markwalter commented that, "Two or three months should be sufficient for standardization," indicating that the standardization process itself should progress quite quickly.
Products from Mid-2010
HDMI Licensing LLC of the US and Blu-ray Disc Association (BDA), among others, hope to see standardization completed as soon as possible, with an eye on product shipment. Support for 3D imagery was written into the HDMI 1.4 standard by HDMI in June 2009, and already companies including Silicon Image Inc of the US and Panasonic have announced HDMI 1.4-compliant receiver ICs and other chipsets.
The BDA created the 3D Task Force in May 2009. According to Masayuki Kozuka, general manager, Storage Devices Business Strategy Office, Panasonic, and BDA co-chair, "We hope BDA will complete standardization before the end of 2009." The reason is that Hollywood is hoping to sell 3D content on Blu-ray Disc format full-scale for the 2010 Christmas shopping season. "In order to be in time," explains Panasonic's Kozuka, "we have to release 3D-capable products in the North American market no later than mid-2010."
Some manufacturers, like Dolby, hope to provide 3D video of fixed quality using Blu-ray Disc and digital videodisc (DVD) player specifications unchanged from what they are now. Panasonic's Kozuka, however, points out that this would lead to a number of problems with both resolution and compatibility: "With current-technology equipment, video resolution would have to be dropped from full high-definition. Except for a special broadcasting where bandwidth restrictions are an issue, we cannot choose a solution that results in lower image quality." In addition, existing specifications include no mechanism for automatically sensing if data is 2D or 3D, and therefore would probably confuse users.
By Tetsuo Nozawa, Nikkei Electronics Asia
