![]() |
Pierre-Alexandre Blanche, Ph.D. Contact
Information |
|
|
3D
display technologies
Abstract What are the different 3D display technologies? What can we expect from them? What are their limitations? How different are they from our holographic system? These are the most frequent questions we get asked about our work. Sony, Cisco, Philips, Toshiba and many others have systems that we will discuss here, to the best of our knowledge, and in layman’s terms.
3D with glasses We humans see 3D mostly because of the stereoscopic vision. Left and right eyes do not see the same image and the brain interprets the difference as the third dimension. I said mostly since there are other factors like the accommodation (strain on the eye lens), atmosphere absorption, cognitive cue... So, any given system that can send different information to the left and right eye can potentially reproduce 3D. And there are a lot of techniques. Let’s discuss the most popular.
Polarization glasses are mostly used nowadays at the theater or by televisions. Each image to be sent to a particular eye is polarized and filtered by the polarizer you are wearing. By changing the polarization every image reaches the correct eye. Shutter eyewear has liquid crystal shutters that are synchronized with the screen. So when the left image is projected, the right shutter is closed and vice versa. The goggles are more cumbersome but apparently the 3D effect is better.
Musion system
The effect, as dramatic as it appears, is only due to the projection on a semi-transparent screen that lets real people and images interact. This technique is known since the XIXth century and referred to as the Pepper's ghost effect. All you need is a good thin screen (what Musion excels at), a projector and some dark background. Cisco 3D telepresence system is based on the exact same technology and once again, no hologram here, no 3D.
Sony 360° 3D display
Actuality-medical has also developed a similar system for medical application that is enclosed in a transparent hemisphere.
CNN holographic interview
Large wall telepresence 3D system
This technology is also used by QinetiQ in another configuration with the projectors on the back of the screen, and by Holografika with the difference that projectors are replaced by a more advanced projection system.
Autostereoscopic television Don't be afraid of the term stereoscopic, it only means the viewer does not have to wear particular glasses to see the 3D effect. Different manufacturers are working on solution: Sony and Toshiba and earlier Philips with its Wow product. How does it work? Remember those lenticular 3D cards where the image changes when tilted? Well, it works the same way but with a dynamic TV screen on the back instead of a printed image. A lenticular array is overlaid on the top of an HDTV, and multiple pixels are recombined together to achieve some 3D effect. It has to be noted that Philips no more commercializes this system because of the inherent flaws of the technology, like false parallax at some positions, and degradation of the resolution from the initial screen.
Star War hologram
So the question is how close can we get?
Generic holographic displays
To give you an order of magnitude, the amount of operation per second it requires is 6 x 1017 flops, nearly one quintillion operations per second (what size and speed)! This is 300 times faster than the current world’s fastest computer, the Jaguar Cray XT5 (2 x 1015 flops.) It will take another 10 years for the fastest computer to reach this computational power according to Moore’s law (fps doubling every 14 months), and it will not hit desktops until the year 2046 (number of transistors double every 2 years). Though, in the future such a technology will certainly be the ultimate holographic 3D display. To compensate for the lack of computational power, Seereal is looking at sub-aperture holograms that are only calculated for the size of the viewer pupil. However, the system needs a head tracking device and only one or two viewers can look at the display at a time.
University of Arizona Holographic Display (see here) Considering the limitation we just discussed above about the computational power, our approach is different. Lasers are used to record the interference pattern into the material and the hologram is decomposed into pixels (or hogels in this case). The images need to be processed before being recorded but nothing like computer generated holograms. It only takes a fraction of a second on a domestic desktop computer to make the calculation. This is why we were able to transmit live images through Ethernet and print them (telepresence). The trade off (nothing comes free) is that we do not reproduce the phase of the scene. So far there were only static holographic recording materials, like photopolymer or silver halide. It means you can write the hologram but not refresh it. It is like a frame in your living room instead of a television. The novelty in our research is that we are using a dynamic material called photorefractive polymer that can be updated at will without any degradation. We can write the hologram, and refresh it over and over again. Our ultimate goal is to write our hologram fast enough so we will have a video rate dynamic holographic display.
|