US 20060238844 A1
A method of computing a hologram by determining the wavefronts at the approximate observer eye position that would be generated by a real version of an object to be reconstructed. In normal computer generated holograms, one determines the wavefronts needed to reconstruct an object; this is not done directly in the present invention. Instead, one determines the wavefronts at an observer window that would be generated by a real object located at the same position of the reconstructed object. One can then back-transform these wavefronts to the hologram to determine how the hologram needs to be encoded to generate these wavefronts. A suitably encoded hologram can then generate a reconstruction of the three-dimensional scene that can be observed by placing one's eyes at the plane of the observer window and looking through the observer window.
1. A method of computing a hologram by determining the wavefronts at the approximate observer eye position that would be generated by a real version of an object to be reconstructed.
2. The method of
3. The method of
3. The method of
4. The method of
a region on the hologram (a) encodes information for that reconstructed point and (b) is the only region in the hologram encoded with information for that point, and (c) is restricted in size to form a portion of the entire hologram, the size being such that multiple reconstructions of that point caused by higher diffraction orders are not visible at the observer eye position.
5. The method of
6. The method of
7. The method of
8. The method of
9. The method of
10. The method of
11. The method of
12. The method of
13. The method of
14. The method of
15. The method of
16. The method of
17. The method of
18. The method of
19. The method of
20. A holographic reconstruction generated from a hologram computed using the method defined in
21. A computer adapted to generate holographic reconstructions using video holograms computed using the method defined in
22. A method of generating a holographic reconstruction of a three dimensional scene using a display device and a computer, the device including a light source and an optical system to illuminate a hologram-bearing medium; comprising the steps of:
(a) using the computer to encode a hologram on the hologram-bearing medium; the hologram having been computed using the method of
(b) illuminating the hologram bearing medium using the light source and optical system so that the reconstructed three dimensional scene is visible.
This application is a divisional application of U.S. patent application Ser. No. 10/534,877 filed by A. Schwerdtner on May 12, 2005 entitled “Video Hologram and Device for Reconstructing Video Holograms”, the contents of which are hereby incorporated by reference.
U.S. patent application Ser. No. 10/534,877 application is, in turn, related to, and claims priority from, PCT patent application PCT/DE03/03791 filed on Nov. 11, 2003 and to German Patent application DE 10253292.3 filed on Nov. 13, 2002, the contents of both of which are hereby incorporated by reference.
The present invention relates to a video hologram and a device for reconstructing video holograms having an optical system that includes at least one light source, a lens and a hologram-bearing medium composed of cells arranged in a matrix or an otherwise regular pattern, with at least one opening per cell, and with the phase or amplitude of the opening being controllable.
Devices for reconstructing video holograms using acousto-optical modulators (AOM) are known from prior art, as detailed in, for instance, U.S. Pat. No. 5,172,251 issued to Benton et al. on Dec. 15, 1992 entitled “Three dimensional display system,” the contents of which are hereby incorporated by reference. Such acousto-optical modulators transform electric signals into optical wave fronts, which are recomposed in a video frame using deflection mirrors to form two-dimensional holographic areas. A scene visible for the viewer is reconstructed from the individual wave fronts using further optical elements. The optical means used, such as lenses and deflection elements, have the dimensions of the reconstructed scenes. Due to their great depth, these elements are voluminous and heavy. It is difficult to miniaturize them, so that their range of applications is limited.
Another way to generate large video holograms is the so-called “tiling method”, using computer-generated holograms (CGH). This method is described in for instance PCT patent publication WO 00/75698 of PCT Patent application PCT/GB2000/001901 filed on May 18, 2000 by Pain et. al entitled “Holographic Displays” and in U.S. Pat. No. 6,437,919 B1 issued to Brown, et al. on Aug. 20, 2002 “System for the production of a dynamic image for display”, both of which are hereby incorporated by reference. In the tiling method, small CGHs having a small pitch are created using an optical system. In the first step of the method, the required information is written to fast matrices that have a small pitch, such as electronically addressable spatial light modulators (EASLM). These fast matrices are then reproduced on to a portion of a suitable holographic medium. A large video hologram is composed of the tiled replicas of the fast matrices. Usually, an optically addressable spatial light modulator (OASLM) is used as holographic medium. In a second step, the composed video hologram is reconstructed with coherent light in transmission or reflection.
In the CGH with controllable openings arranged in a matrix or in an otherwise regular pattern as described in, for instance, PCT publications number WO 01/95016 A1 of PCT patent application PCT/GB2001/002302 filed on May 24, 2001 by Payne et al. entitled “Computation Time Reduction for Three-Dimensional Displays” or in Fukaya et al., “Eye-position tracking type electro-holographic display using liquid crystal devices”, Proceedings of EOS Topical Meeting on Diffractive Optics, 1997, the diffraction on small openings is taken advantage of for encoding the scenes. The wave fronts emerging from the openings converge in object points of the three-dimensional scene before they reach the viewer. The smaller the pitch, and thus the smaller the openings in the CGHs, the greater is the diffraction angle, i.e. the viewing angle. Consequently, with these known methods enlarging the viewing angle means to improve the resolution.
As is generally known, in Fourier holograms the scene is reconstructed as a direct or inverse Fourier transform of the hologram in a plane. This reconstruction is continued periodically at a periodicity interval, the extension of the periodicity interval being inversely proportional to the pitch in the hologram.
If the dimension of the reconstruction of the Fourier hologram exceeds the periodicity interval, adjacent diffraction orders will overlap. As the resolution is gradually decreased, i.e. as the pitch of the openings increases, the edges of the reconstruction will be increasingly distorted by overlapping higher diffraction orders. The usable extent of the reconstruction is thus gradually limited.
If greater periodicity intervals and thus greater viewing angles are to be achieved, the required pitch in the hologram comes closer to the wavelength of the light. Then, the CGHs must be sufficiently large in order to be able to reconstruct large scenes. These two conditions require a large CGH having a great number of openings. However, this is currently not feasible in the form of displays with controllable openings, as discussed in, for instance, U.S. Pat. No. 6,831,678 issued to Travis on Dec. 14, 2004 entitled “Autostereoscopic display”, the contents of which are hereby incorporated by reference. CGH with controllable openings only measure one to several inches, with the pitches still being substantially greater than 1 μm.
The two parameters, pitch and hologram size, are characterized by the so-called space-bandwidth product (SBP) as the number of openings in the hologram. If the reconstruction of a CGH with controllable openings that has a width of 50 cm is to be generated so that a viewer can see the scene at a distance of 1 m and in a 50-cm-wide horizontal viewing window, the SBP in horizontal direction is about 0.5×106. This corresponds to 500,000 openings at a distance of 1 μm in the CGH. Assuming an aspect ratio of 4:3, 375,000 openings are required in the vertical direction. Consequently, the CGH comprises 3.75×1011 openings, if three color sub-pixels are taken into consideration. This number will be tripled if the fact is taken into account that the CGH with controllable openings usually only allows the amplitudes to be affected. The phases are encoded taking advantage of the so-called detour phase effect, which requires at least three equidistant openings per sampling point. SLM having such a great number of controllable openings are hitherto unknown.
The hologram values must be calculated from the scenes to be reconstructed. Assuming a color depth of 1 Byte for each of the three primary colors and a frame rate of 50 Hz, a CGH requires an information flow rate of 50*1012=0.5*1014 Byte/s. Fourier transformations of data flows of this magnitude exceed the capabilities of today's computers by far and do thus not allow holograms to be calculated based on local computers. However, transmitting such an amount of data through data networks is presently unfeasible for normal users.
In order to reduce the enormous number of computations it has been proposed not to calculate the entire hologram, but only such parts of it that can be seen directly by the viewer, or such parts that change. The kind of hologram which consists of addressable sub-regions, such as the above-mentioned “tiling hologram”, is disclosed in the above-mentioned patent specification WO 01/95016 A1. Starting point of the calculations is a so-called effective exit pupil, the position of which can coincide with the eye pupil of the viewer. The image is tracked as the viewer position changes by continuous recalculation of the hologram part that generates the image for the new viewer position. However, this partly nullifies the reduction in the number of computations.
The disadvantages of the known methods can be summarized as follows: Arrangements with acousto-optical modulators are too voluminous and cannot be reduced to dimensions known from state-of-the-art flat displays; video holograms generated using the tiling method are two-stage processes which require enormous technical efforts and which cannot easily be reduced to desktop dimensions; and arrangements based on SLM with controllable openings are too small to be able to reconstruct large scenes. There are currently no large controllable SLM with extremely small pitches, which would be needed for this, and this technology is further limited by the computer performance and data network bandwidth available today.
The invention is defined in Claim 1. In one implementation, video holograms and devices for reconstructing video holograms with controllable openings according to the present invention are characterized in that in the viewing plane at least one viewing window is formed in a periodicity interval as a direct or inverse Fourier transform of the video hologram, said viewing window allowing a viewer to view a reconstruction of a three-dimensional scene. The maximal extent of the viewing window corresponds to the periodicity interval in the plane of the inverse Fourier transformation in the image plane of the light source. A frustum stretches between the hologram and the viewing window. This frustum contains the entire three-dimensional scene as a Fresnel transform of the video hologram.
The viewing window is limited approximately to and positioned in relation to one eye, an eye distance of a viewer or to another suitable area.
In an implementation, another viewing window is provided for the other eye of the viewer. This is achieved by the fact that the observed light source is displaced or added a second, real or virtual, adequately coherent light source at another suitable position to form a pair of light sources in the optical system. This arrangement allows the three-dimensional scene to be seen with both eyes through two associated viewing windows. The content of the video hologram can be changed, i.e. re-encoded, according to the eye position in synchronism with the activation of the second viewing window. If several viewers view the scene, more viewing windows can be generated by turning on additional light sources.
In another implementation of the device for reconstructing a video hologram, the optical system and the hologram-bearing medium are arranged so that the higher diffraction orders of the video hologram have a zero point for the first viewing window or an intensity minimum at the position of the second viewing window. This prevents the viewing window for one eye to cross-talk the other eye of the viewer or to other viewers. It is thus taken advantage of the decrease in intensity of the light towards higher diffraction orders, which is due to the finite width of the openings of the hologram-bearing medium and/or the minima of the intensity distribution. The intensity distribution for rectangular openings, for example, is a sinc2 function which rapidly decreases in amplitude and forms a sin2 function which decreases as the distance grows.
The number of openings in the display determines the maximum number of values that must be calculated for the video hologram. The transmission of data from a computer or through a network to the display representing the video hologram is limited to the same number of values. The data flow rate does not substantially differ from the data flow rates known from typical displays used today. Now, this will be illustrated with the help of an example.
If the viewing window is reduced, for example, from 50 cm (horizontal) by 37.5 cm (vertical) to 1 cm by 1 cm by choosing a sufficiently low-resolution display, the number of openings in the hologram will drop to 1/1875. The required bandwidth is reduced in the same way during data transmission through a network. Video holograms created with known methods require 1012 openings, while this number is reduced to 5108 pixels in this example. The scene can be viewed in full through the remaining viewing window. These requirements on pitch and hologram size according to the space-bandwidth product can already be fulfilled by displays available today. This allows the inexpensive realization of large real-time video holograms on displays with large pitches and having a large viewing window.
The viewing window is tracked by mechanically or electronically displacing the light sources, by using movable mirrors or by using light sources which can be adequately positioned in any other way. The viewing windows are displaced according to the displacement of the light source images. If the viewer moves, the light source(s) is (are) spatially displaced so that the viewing windows follow the eyes of the viewer(s). This is to ensure that the viewers can also see the reconstructed three-dimensional scene when they move, so that their freedom of movement is not limited. Several systems are known for detecting the position of the viewers, e.g. systems based on magnetic sensors can be used beneficially for this.
An implementation of this invention also allows the efficient reconstruction of a video hologram in color. Here, the reconstruction is performed with at least three openings per cell, representing the three primary colors. The amplitude or phase of the openings may be controllable, and the openings may be encoded individually for each of the primary colors. Another possibility of reconstructing a video hologram in color is to perform at least three reconstructions one after another, namely for the individual primary colors, using the device of the present invention.
An implementation of this invention allows the efficient generation of holographic reconstructions of spatially extended scenes through controllable displays, such as TFT flat screens, in real-time and providing large viewing angles. These video holograms can be used beneficially in TV, multimedia, game and design applications, in the medical and military sectors, and in many other areas of economy and society. The three-dimensional scenes can be generated by a computer or in any other way.
These and other features of the invention will be more fully understood by references to the following drawings.
An embodiment of the present invention is illustrated and explained below in conjunction with the accompanying drawings, wherein
A device for reconstructing video holograms comprises the hologram-bearing medium, a sufficiently coherent, real or virtual, point or line light source and an optical system. The video hologram-bearing medium itself consists of cells which are arranged in a matrix or in an otherwise regular pattern with at least one opening per cell, the phase or amplitude of said opening being controllable. The optical system for reconstructing the video hologram can be realized by an optical imaging system known in the art, consisting of a point or line laser or a sufficiently coherent light source.
The light source 1 is imaged on to the viewing plane 4 through an optical system, represented by the lens 2. If a hologram-bearing medium 3 is inserted, it is reconstructed in the viewing plane 4 as an inverse Fourier transform. The hologram-bearing medium 3 with periodic openings creates equidistantly staggered diffraction orders in the viewing plane 4, where the holographic encoding into higher diffraction orders takes place, e.g. by way of the so-called detour phase effect. Because the light intensity decreases towards higher diffraction orders, the 1st or −1st diffraction order is used as the viewing window 5. If not explicitly expressed otherwise, the 1st diffraction order will be taken as a basis in the further description of the invention.
The dimension of the reconstruction was chosen here to correspond with the dimension of the periodicity interval of the 1st diffraction order in the viewing plane 4. Consequently, higher diffraction orders are attached without forming a gap, but also without overlapping.
Being the Fourier transform, the selected 1st diffraction order forms the reconstruction of the hologram-bearing medium 3. However, it does not represent the actual three-dimensional scene 6. It is only used as the viewing window 5 through which the three-dimensional scene 6 can be observed (see
The intensity decrease towards higher diffraction orders is taken advantage of to suppress cross-talking to other viewing windows.
Of course, the viewer can view the scene 6 of the hologram 3 with both eyes (see
For the left eye, the 1st diffraction order corresponding to the position of the light source 1 was chosen. The left viewing window 5 is formed likewise. According to an implementation of this invention, the corresponding three-dimensional scenes 6 and 6′ (not shown) are reconstructed using the light sources 1 and 1′ in a fix position in relation to the eyes. For this, the hologram 3 will be re-encoded when the light sources 1 and 1′ are turned on. Alternatively, the two light sources, 1 and 1′, can simultaneously reconstruct the hologram 3 in the two viewing windows 5 and 5′.
If the viewer moves, the light sources 1 and 1′ are tracked so that the two viewing windows 5 and 5′ remain localized on the eyes of the viewer. The same applies for movements in the normal direction, i.e. perpendicular to the video hologram.
Further, several viewers can view a three-dimensional scene if additional viewing windows are created by turning on additional light sources.
Although the invention has been described in language specific to structural features and/or methodological acts, it is to be understood that the invention defined in the appended claims is not necessarily limited to the specific features or acts described. Rather, the specific features and acts are disclosed as exemplary forms of implementing the claimed invention. Modifications may readily be devised by those ordinarily skilled in the art without departing from the spirit or scope of the present invention.