Introduction

Integral photography is a method of recording and recreating three-dimensional scenes. In the future, integral photography, may form the basis for 3D television. Research prototypes of such systems are already being built in Japan by NHK. Display holography and integral photography are competing technologies for this purpose.

Principles of Operation

Central to the technique of integral photography is a sheet of tiny hemispherical lenslets arranged in a honeycomb pattern. At each tiny lenslet, there are light rays coming to the center of the lenslet from all directions in the hemisphere. The focal length of the lenslet is set so that each ray (from a different direction) focuses on a different point on the flat circular base of the lens.

The image on the base of the sheet is then recorded (ex. with a CCD array or using conventional film). The film below each circular base records the light coming from an entire hemisphere of directions into the center of the base. Adjacent lenslets in the honeycomb image the hemisphere centered one their respective base centers. All together, the lenslets allow us to get a complete (although rough) sample of the total-light function (plenoptic function) of the 3D scene.

To reconstruct the scene, one merely reverses the process. The recorded image is projected through the bottom of the sheet (ex. with a LCD). Alternatively, a synthetic pattern can be computed using computer graphics techniques and then projected. In either case, the viewer will be able to see the 3D scene with out any viewing aids, like 3D goggles. This type of 3D display device is called an autostereoscopic display.

Complexity

Obviously, the smaller and closer together the lenslets, the better the resolution. Lenslet sheets with tiny lenslets can be produced using photolithographic techniques. Cylindrical islands of resist are melted so that surface tension forces them into a hemispheres. However, making lenslet sheets with large area is a different and difficult problem.

We need smaller lenslets not only to achieve higher linear spatial resolution. With high enough resolution, we should be able to fool the human physiological depth cue of accommodation. However, integral photography, like any other technique in geometrical optics, is limited by diffraction so there is a limit to how small we can make the lenslets.

Aside from theoretical limitations, the current state of the art of integral photography is limited mostly by the resolutions of the CCD and LCD arrays, and by the sheer amount of information required to process a 3D scene.

Simplifications

The complexity of integral photography has prompted the development of a simplification. In it, hemi-cylindrical lenslets are used instead of hemispherical ones. Such sheets of hemi-cylindrical lenslets are called lenticular arrays. These sheets are often adhered to a static image pattern and used to create "3D" baseball cards and bookmarks. You should be able to feel the texture of the hemi-cylindrical lenslets on such baseball cards.

Because of the simplification lenticular sheets support only horizonal parallax, and not vertical parallax. In the case of the baseball cards, this means that the image changes as the angle of view is varied horizontally but not vertically.

Log in or registerto write something here or to contact authors.