Recent remarkable progress in the field of media generation, media representation, media compression, media display, and the larger and ever-increasing bandwidth available to the customer will foster the introduction of immersive media services in the near future. It is widely accepted that this trend towards immersive media is going to have a strong impact on our daily life. The ability to evoke a state of being there and of being immersive into media applications will no longer remain in the domain of compute simulators, virtual reality and CAVE systems. It will arise in offices, venues, and homes and it has the potential to enhance quality of life in general.

3D information display to create a more lifelike 3D image is regarded as the core technology for these immersive multimedia telecommunications. It is well known that two types of 3D technology are now employed in the practical application fields. One is the shading and rendering techniques mostly used in graphic design, games and many applications to give the illusion of 3D on a 2D image, which is basically classified as 2D technology because rendered 2D images are displayed on 2D displays. The other is the display technologies to create more lifelike 3D images on 3D displays, which is actually known as real 3D technology. These include stereoscopic, autostereoscopic, volumetric, spatial and holographic 3D technologies.

Here in this series column we mainly concern on these real 3D technologies. Believe it or not, research on 3D photography have a long history of over 150 years. In 1838, Wheatstone of England established that humans perceive depth when the brain combines two slightly different images. Each eye, separated by about sixty-five millimeters, sees things with a slightly different aspect. Wheatstone believed that this discrepancy, known as a stereo disparity, allows us to see the world in three dimensions. He invented an instrument he called the Stereoscope and demonstrated a 3D display. Within fifty years of that moment, stereo photography grew to become one of the most popular forms of entertainment, becoming as widespread as TV is today. In the 1890's nearly every home owned a stereoscope. The popularity of this stereo photography declined rapidly after the introduction of the Kodak Brownie camera in 1900.

Within another fifty years however, 3D became popular entertainment again. The invention of Kodak chrome color film as well as the introduction of the Realist stereo camera stimulated a renewed interest in personal, color 3D photography, while everything from Viewmaster discs, comic books, and a variety of movies were all produced in 3D during the 1950's. In the 1960's the arrival of the Kodak Instamatic camera together with the color Polaroid hold backed interests in 3D photography again. Recently 3D has started growing once again. The widespread adoption of personal computers, digital cameras and desktop photo printers has now removed any obstacles to stereo 3D photography in the home or office.

With the introduction of Photo 3D, now anyone who can operate a digital camera can take their own, eye-popping 3D photos.

So far, there are more than 50 kinds of 3D displays developed up to now. They can be largely categorized as stereoscopic and autostereoscopic 3D displays depending on image separation methods. In the stereoscopic 3D display, the viewer is required to wear special glasses such as polarized or shutter glasses for separated reception of the left and right images by the eyes. But autostereoscopic 3D displays can present a 3D image to the viewer without a need for any special glasses. Optical elements such as lenticular sheets or parallax barriers are attached to the display panels and they make the left and right images displayed on the panel to be collected on the corresponding eyes without any interference between them.

Meanwhile a holographic 3D display, which is totally different from the conventional stereoscopic and autostereoscopic approaches, has been regarded as one of the most attractive approaches for creating the most authentic illusion of observing volumetric objects. This is because the holographic technology can supply high-quality images and accurate depth cues viewed by human eyes without any special observation devices.

Integral imaging known as a volumetric 3D display is also one of the promising 3D technologies because it can provide full-color, full parallax and continuous viewing point images without a need of special glasses. In addition, the Fresnel lens-based spatial 3D display system has been also suggested, in which 3D images with real depth is floated into the air by reflecting an object image through one or more mirrors and lenses. These holographic, volumetric and spatial 3D displays mentioned above will be covered in the next column and stereoscopic and autostereoscopic 3D displays are discussed here in this first column.

Generally, human beings see different viewpoint images of an object through two eyes, the right and left. Then, the human brain recognizes the three-dimensional (3D) stereopsis of the object by synthesizing them with the binocular disparity of a stereo input image pair. Most conventional 3D display systems have been implemented by imitating this human visual system, so that these systems normally need a pair of cameras for capturing the left and right images of an object, and a pair of projectors for projecting the captured left and right images to the screen, and some special optical devices for separately inputting the projected left and right images to the corresponding eyes.

Many older stereoscopic 3D displays were based on the high imagequality and fast response of CRTs. But as high-quality and very narrow flat panel displays such as LCDs, PDPs, and OELED are replacing the markets of the conventional CRTs, R&D into the use of LCDs for 3D applications is driven. This R&D has been fruitful and has produced three technologies that can produce 3D image quality comparable to 2D image quality on LCD displays. The first one is the two LCD systems where the viewer looks at a half silvered mirror which superimposes the images from the two LCDs. Passive polarizing glasses then separate these two images. Planar is known as one of the leading developers of this technology.

Second is the dual-LCD systems where two LCDs are stacked, one controlling the brightness and the other controlling the left/right separation. The viewer wears passive polarizing glasses to separate the images.

NuVision is developing products based on this technology. The third one is a single micropolarized LCD system in which alternate rows in a LCD display have different polarization. The viewer wears passive polarizing glasses. The technology was developed by VRex and has been turned into products by customers of Arisawa. It is noted here that these 3D LCD systems can be also used for 2D displays if the user removes his glasses and provides a 2D video signal to these displays. These displays have all the properties of the underlying 2D LCDs, including color, brightness, resolution, and contrast.

In the stereoscopic 3D displays discussed above, they require viewers to wear special glasses such as a pair of polarized or shuttered glasses to feel the depth, which has been regarded as the main shortcoming in practical applications. Therefore, many kinds of autostereoscopic displays have been developed, because autostereoscopic displays could present a 3D image to viewers without the need for any special glasses.

The most common glasses-free technology is autostereoscopic LCD systems and there are two flavors of this technology, parallax barrier and lenticular sheet. In the parallax barrier approach, left and right eye perspective views of the image are created using separate pixels on the LCD display. In a two view display for example, the right eye perspective image is presented on an LCD panel display utilizing only the odd pixel groups of columns and the left eye perspective image is presented on the even pixel columns. In doing this, the horizontal resolution of the image has been reduced by half compared to the normal display resolution.

A parallax barrier sheet is placed in front of the LCD display. It is a clear sheet containing a series of narrow, linear, opaque stripes. The width and spacing of the stripes and the position of the sheet with respect to the width and spacing of the pixel columns in the display is such that a viewer looking at the display is able to see the right eye pixel columns from some viewing angles but not from others. The same is true for the left eye pixel columns. Since each eye sees a different image, if the images are a stereoscopic pair, the viewer sees it in 3D. Moreover, an autostereoscopic effect can also be achieved with a lenticular sheet. This type of sheet contains a linear array of narrow, cylindrical lenses. Their function is to direct light from the image to different areas in the viewing zone. This image is greatly distorted because a lenslet would cover several groups of columns; 2 in a 2-view display or 9 in a nine view display. Nonetheless, it shows the fact that no light is blocked by the lenslets; all light is directed into the viewer space. This means that an autostereoscopic display based on lenslets does not have the brightness problem associated with parallax barriers.

Two-view systems are generally used only for applications where there will only be a single user. Multi-view applications are typically designed to present 8.25 views and be seen by a larger audience. The lenticular sheet or parallax barrier directs the light so from two different spots you see two different images. These spots are about 65 millimeters apart, the normal separation between human eyes. Twoview produces only one pair of images and must be viewed from a certain location typically on the centerline of the display. Multi-view displays have a larger number of stereo pairs and can be viewed from multiple locations.