HOLOGRAPHIC TV
 

-1997-

Objective:

The objective is to support DARPA as it seeks to develop a three dimensional holographic conferencing technology. In particular this effort will explore the technology of diffractive optics as it may be applied to the goal of three dimensional image projection.

Approach:

In this approach images are projected holographically using dynamic diffractive optical elements. Software developed by the diffractive optics industry is used to create a diffraction grating which represents a two dimensional or three dimensional image. This diffraction grating pattern is formed on a phase Spatial Light Modulator, SLM, or other phase modulating media. When laser light of the correct color is transmitted through this grating a two or three dimensional image is generated in the farfield. This image results from the diffraction of a coherent laser light through the computer generated two dimensional diffraction grating. Since holograms create the images from interference effects, the system requires no optics or focus adjustments. The images are observed when the diffracted light scatters from a surface such as a wall or a scattering volume. The projection device can be small and compact as it consists of only two components, a phase SLM and a laser. The images formed can be quite large and do not require that the observer look into a box. The complexity of the image is limited by the bandwidth of the two dimensional phase SLM.

A three dimensional image can be formed using stereoscopic image projection. A two SLM system operating in parallel will function as a stereoscopic projection system. The first SLM projects an image as observed from the left eye and the second SLM projected an image as observed from the right eye. Each image is projected at either a different polarization or a different wavelength. The observer is fitted with special glasses with either polarization filters or notched color filters. The observer will experience the reality of a three dimensional image. The polarization approach can provide multiple viewers with a three dimensional image however each viewer will observe the same perspective. The multiple wavelength approach can be extended to provide multiple perspectives to multiple observers. In this concept each observer at each perspective is assigned two colors or wavelengths one for the left eye and one for the right. Each observer wears glasses or contacts with color filters passing only his color bands. A two SLM projection system is used for each observer prospective. Each system using laser wavelengths appropriate for that observer. The array of two SLM projector's are used to project two images, right and left, for each observer located at specific perspectives. All observers in the room observe a three dimensional image, however, each observer experiences a different experience appropriate to his location or perspective from the projected three dimensional object. This approach has the advantage of minimizing system bandwidth and processing only the minimum information necessary to create the three dimensional experience.

This project will demonstrate the use of diffraction to project video images. The use of diffraction for projection has several advantages over conventional image projection. In conventional video projection the image projected has a finite depth of field and must be carefully focused to a given distance. The projection system uses a large objective lens. The entire projector is large and bulky. In contrast, diffraction projection uses a very small phase modulation device to diffract light into forming the desired image. The image is formed by diffraction rather than projection. The image has infinite depth of field without a large projection lens. The diffraction based system requires no focus adjustment, the images are always in focus at any distance. The diffraction system is much simpler and very compact. To accomplish this innovation video images must be converted into diffraction patterns in real time. SY Technology, Inc. has developed highly specialized software to convert video images into diffraction patterns. One goal of this project is increase the speed of this computation to as near real time as possible.

Diffraction projection has many advantages for stereoscopic display. In particular the images can be formed at any distance without the need for range focus adjustment. Also as the system uses laser light, color multiplexing can be used to provide independent perspective views for different observers.
 

Recent FY-97 Accomplishments:

New Start

FY 98 Plans
 

Optimize design software to run as close to real time as possible. This task will require numerical methods and specialize hardware to parallelize the algorithms.

Use this software to design the diffractive phase patterns required to generate a short three dimensional, two perspective projected movie.

Build and demonstrate a three dimensional, two perspective projected movie using diffractive optics.


-1998-

Objective:

The long-range goal of this program of research is to support the Defense Advanced Research Projects Administration, DARPA, as it seeks to develop a three-dimensional (3D) holographic conferencing technology. In particular this effort will explore the technology of diffractive optics as it may be applied to the goal of 3D image projection. The motivation for this research is that the present technology of telephone and video conferencing has proven inadequate in providing the teleconferencing participants the experience, look and feel that accompanies a real meeting between human beings. The loss of this feeling of a real human interaction can effect how humans interact. A participant of a remote teleconference often feels detached and anonymous from the other participants. The bonding and human association between participants is often lost. As a consequence of this the out come of conferences, negotiations, treaties, contracts, battle plans etc. are substantially different. The technologies employed to accomplish this task are novel autostereoscopic display and stereoscopic projection systems with diffractive and refractive micro-optic components.

Approach:

The approach taken in the research program is experimental in nature. The desire is to transfer technology from the diffractive optics industry into the holographic display market. To facilitate this technology transfer, two technologies are examined in the program: an autostereoscopic display interfaced to a camera that records 3D images autosterescopically, and a stereoscopic projection system using dynamic diffractive optic elements (DOE).
In the first approach, it is desired to form a three dimensional image using a flat-panel display with affixed refractive micro-optics. The micro-optics essentially sub-divide the display picture elements (pixels) optically, so that information encoded on the individual pixel subsets are viewed separately by each eye. This is the essence of autostereoscopy: various images are projected to each eye, with no encumbrances on the viewer other than his location, and the human psycho-visual system integrates the separate views into one 3D image. With the proper configuration of micro-lenses and display picture elements, more than two views can be displayed to the viewer. Thus, as the viewer moves his head laterally at some distance from the display, he will have more than two 3D depth cues and 3D perception will be obtained with greater ease. The simple flat-panel autostereoscopic display functions in color or grayscale with no special changes to the laptop other than the addition of the micro-optics.

The flat-panel display is complemented in the proposed research by a camera that records data autosterescopically. The camera essentially performs the inverse operation of the display. The camera contains micro-optics attached to the detector array which optically subdivide the recorded video image in such a way that it can be redisplayed on the flat-panel display with no additional hardware or optics other than those mentioned above. Formats for the camera and display micro-optics are the same. Thus, the camera and the display form a closed loop system for real-time display and recording of autostereoscopic 3D images.

The second approach demonstrates the use of diffraction to project conventional video images and 3D images. Conventional video projectors are large and bulky. The images produced have finite depth of field and require careful focusing. Diffracted images are created when laser light of the correct color or wavelength passes through a diffraction grating. The grating is formed on an electronically addressable Spatial Light Modulator, SLM, which modulates the phase, or directional character, of incident light. Thus, the SLM functions as a dynamic DOE and diffracted images can be formed in real-time. The image is viewed when the diffracted light scatters from a diffuse surface or scattering volume placed in the far-field of the SLM. Thus, no large projection/focusing optics are required and the image has infinite depth of field and is always in focus. Since the projection system consists of a laser and an SLM, it is simple and compact. To accomplish this innovation video images must be converted into diffraction patterns in real time. SY Technology, Inc. has developed highly specialized software to convert video images into diffraction patterns. One goal of this project is increase the speed of this computation to as near real time as possible.

The 3D projection system forms a stereoscopic image pair using two SLMs operating in parallel. One SLM projects an image as observed from the left eye and the other SLM projects an image as observed from the right eye. Each image is projected at either a different polarization or a different wavelength. The observer is fitted with special glasses with either polarization filters or notched color filters. The observer will experience the reality of a large 3D image without looking into a box. The complexity of the image is limited by the bandwidth of the two-dimensional (2D) phase SLM.

This concept is extended to multiple views or multiple viewers by assigning a separate polarization, or wavelength, to each eye. A two SLM projection system is used for each observer in the room unless it is desired that all see the same perspectives. All observers in the room observe a 3D image, however, each observer experiences a different experience appropriate to his location or perspective from the projected three dimensional object. This approach has the advantage of minimizing system bandwidth and processing only the minimum information necessary to create the 3D experience.

Diffraction projection has many advantages for stereoscopic display. In particular the images can be formed at any distance without the need for range focus adjustment. Also as the system uses laser light, color or polarization multiplexing can be used to provide independent perspective views for different observers.
 

Recent FY-97 Accomplishments:

Developed a software algorithm that calculates view-interlacing requirements and interlaces multiple object views. This software functions in the place of the autostereoscopic camera for training or prototyping purposes.
Designed and fabricated micro-optic lenslet arrays that attach to the flat-panel display. The optics reverse the interlacing process and direct the image views to the appropriate eye without the need for headgear. The display will have application to remote vehicle piloting for mine clearing, aim point designation, destructive and nondestructive testing, treaty negotiation, battlefield visualization, and medical training.

Developed the optical design for the autostereoscopic camera. This camera is designed to complement the autostereoscopic flat-panel display, and it evolves as the display evolves to incorporate a multi-view architecture. While the camera design is specialized, the sensor or detector array is not. Therefore, the camera has application to other sensor requirements such as night vision. The autostereoscopic camera also has potential application to ATR by adding an extra tuning variable.

Designed and implemented a dynamic DOE design algorithm in a DSP image processing hardware architecture. The DSP generates a diffractive phase mask from a conventional camera image using a modified Gerschberg-Saxton algorithm, and forms the image on a phase SLM.

Designed the optical hardware to implement a single phase SLM as a dynamic DOE-based projector. Other than demagnification optics for the SLM image, which increases the usable projection solid angle, there are no actual projection optics. Thus, the hardware can be miniaturized for field operations.
 

FY-98 Plans:
 

Update the flat-panel display medium to incorporate more than two object views. Examine alternate optical image compression schemes to incorporate multiple views. Examine diffractive color filter designs to eliminate chromatic dispersion and increase the number of usable horizontal pixels.
Conduct trade studies of the autostereoscopic camera design to extend the field-of-view (FOV), the viewer location space, and the number of views recorded to complement the autostereoscopic display.

Optimize DOE design software and hardware to run as close to real time as possible. This may require alternate DSP strategies, such as FPGAs (field programmable gate arrays).

Design algorithms to adjust SLM phase modulating effects in real-time for color adaptable displays. Incorporate planarized-pixel SLMs and collimating lenslets into projector design to increase optical throughput. Build and test stereoscopic projector using two phase SLM design, with a focus on miniaturization and extension to multiple views.

Technology Transition:

The Holographic Television program is in the prototyping and basic research stages for both the autostereoscopic display and camera and the dynamic DOE-based stereoscopic projector. However, demonstrations of the prototype experimental architectures are scheduled.
The technology developed under this program will be licensed to MEMS Optical, Inc., a subsidiary formed to market SY Technology's commercializable technology, and any other partners necessary to bring these products to the commercial market. For example, the laptop display market will be the first obvious large demand source for the flat-panel autostereoscopic display. Once demonstrated, it is expected that the product technology will attract enough military interest to go into continued R&D as well as commercial production of some models.