The present invention relates generally to an apparatus that enables 3D displaying.
Today's developed displays with advanced technologies including “Liquid Crystal Display (LCD)” show images with very high quality. However, there is a vital inadequacy with today's 2D displays. This inadequacy is a result of expressing the 3D real world on a 2D plane and ignoring the fact that human beings experience the real world through two different eyes. In vision of the real world, two eyes correspond to two different views for the visual system while traditional displays provide only one view to the visual system—the same view towards each eye. 3D Displays seem to be the next step in the evolution of displays and will overcome this inadequacy by providing different views to different eyes. With the incredible developments in the digital video processing and visualizing technologies, first commercial 3D display products are already available in the market. It is helpful to classify 3D displays for a better understanding of their development trend and a possible classification can be as holographic displays, volumetric displays and auto-stereoscopic displays [1].
First group, holographic displays in spite of their great potential stemming from their 3D reconstruction quality, are not strong candidates for being widespread and commercial in the following years, due to high bandwidth requirements, demand for SLMs with high resolution and difficulties in achieving natural shading. Second group, volumetric displays have a variety of approaches e.g. real image methods applying static or moving displays and few commercial products realizing these approaches. Perspecta developed by Actuality Systems having a rotating disc at 900 rpm on which images are projected sequentially is a good example for real image approaches with a moving display. Fogscreen, creating an image on fog-like particles which seems to be floating in the air is a good example for real image approaches with static displays. The volumetric displays have the vital drawback of transparency. It means that objects that should be behind some other objects, are not occluded by the front object and seen by the viewer which cause a confliction in viewer's 3D perception. Another drawback with volumetric displays is their incapability of displaying surfaces having non-Lambertian intensity distributions. Today, the third group, auto stereoscopic multi-view displays e.g. Philips' multi-view display using slanted lenticular sheet or Sanyo's multi-view display using parallax barrier, seem to have the highest potential of acceptance in the display market in the following years. However, auto stereoscopic displays have also their own drawbacks including: generation of pseudoscopic viewing regions, decrease in resolution with increasing view number, discontinuities and jumps between adjacent views, eye fatigue stemming from accordance problem of accommodation and vergence mechanisms of the eye.
Holographic-like displays solve some major problems of the auto stereoscopic displays mentioned above and provide the key advantages of holographic displays such as accommodation-vergence synchronization and smoother motion parallax by constituting larger number of views in the field of view [1], [2]. Actually, it is found that twenty views per interocular distance is an optimum value for smooth motion parallax. There are a few examples of holographic-like displays that use micro display array and collimated light source [3], [4], a laser or array of laser diodes and 2D scanners [5], [6].
U.S. Pat. No. 6,999,071 issued in February 2006 explains such a 3D Display method. The 3D display D3D tries to realize a 2D Screen 20 with screen pixels S that can emit light with different colors and intensities to different directions LS1 to LSn[3]-[6]. The system transmits independently modulated light beams LM in different directions LS1 to LSn from a single screen point S in contrast to traditional 2D displays D2D transmitting the same light information in every direction from a single screen point S as illustrated in
This is accomplished by illuminating numerous 2D micro displays 60 controlled according to the 3D image that will be displayed. The light from the light source 13 is collimated before illuminating the micro displays 60. Each independently modulated light beam LM by the individual pixels S of the 2D micro displays is then transmitted in different directions by a lens system 31 and 32 present in front of each 2D micro display 60, as shown in
In this invention, the above 3D visualization concept, approaching 3D displays as 2D displays that have pixels emitting different color and intensity light to different directions, is realized by using an array of scanners that images properly modulated light to the proper screen pixels on their scanning path.
In a preferred embodiment of the system, 1D array of light sources per each main color are integrated with 1D modules scanning in torsion mode together with imaging lenses. The light sources are modulated by a driving circuitry which is mounted ON or OFF the scanning platform. There is 2D array of these scanning modules behind the screen placed with a specific periodicity to a specific distance according to the resolution requirements of the display and the number of different views the display requires to provide. The precisely controlled intersections of rays coming from several scanning modules correspond to a complete set of voxels and the viewers looking from different perspectives will see different 3D images. In the system, light sources are preferably LEDs or organic LEDs and scanners are preferably made from polymer or silicon materials.
Another preferable scanning mode can be in-plane mode but in this mode the imaging lens will not be connected to the scanning platform. The module will scan behind a motionless lens and according to scanner's relative position to the lens; the ray bundles emitted from the light sources on the scanner will be directed to different screen pixels.
In a further advantageous implementation, the light sources can be motionless and the lens is scanned in in-plane mode in front of the light sources to image them to different screen pixels.
Different actuation mechanisms such as electrostatic or electromagnetic actuation can be used for realizing the scanning. In a preferred system, electromagnetic actuation with a magnet placed on top of the scanner interacts with an external electro coil driven with alternating current. In a further preferred system, the electro coil can be printed or fabricated on to the scanner and actuation can be realized by an external magnet.
In another implementation of the system, instead of using 1D array of light sources per each main color coupled with 1D scanner, a single light source per each main color coupled with 2D scanners is used. Here the light source can be preferably laser diodes, vertical cavity surface emitting diodes (VCSELs). Scanners are preferably made from polymer or silicon materials or from both of them. The light sources can be on top of the scanners or they can be external and their light can be reflected to the screen pixels by a mirror placed on top of the 2D scanners.
In all configurations, the scanning angle of the scanners can be limited to a specific narrow angle with a specific offset if only limited numbers of viewers are viewing the display from a limited viewing angle. This embodiment of the system is quiet advantageous as it will increase the efficiency and as a result brightness of the display.
In another system, a special screen that can move left and right directions according to the position of the viewers constituted from an array of cylindrical lenses that have modulatable pitch sizes can be used together with a head tracking system to send 3D information only to the specific region where viewers are standing. This system can be preferably used with personal devices. This special screen can be used either in front of displays having light sources located at the pixel positions including liquid crystal displays (LCD) or displays that have pixels scanned with at least one scanner coupled with at least one light source in a certain depth.
FIG. 1—The working principle of the quasi-holographic volumetric display
FIG. 2—The basic unit of the Holografika display
FIG. 3—Different viewers looking from different perspectives receive different views
FIG. 4—(a) 1D LED arrays in RGB colors and driver IC mounted on FR4 scanner platform; (b) Scanner modules as the basic unit of the 3D display.
FIG. 5—Every pixel on the screen is illuminated by different modules whose number is equal to the number of different emission directions from the pixel
FIG. 6—Voxels rendered (i) in front of the screen, (ii) between the screen and the LED modules, (iii) behind the LED modules.
FIG. 7—The optical behavior of the system in vertical and horizontal directions.
FIG. 8—Micro lens array in superposition mode to image the light sources onto the screen
FIG. 9—Micro lens array in apposition mode to image the light sources onto the screen
FIG. 10—3D display scanning modules implementation with lateral translations of a lens
FIG. 11—FPGA as a LED Driver on polymer scanner for driving the LED array
FIG. 12—The complete display
FIG. 13—2D Scanning Based 3D Display Concept using laser diodes placed on top of polymer scanners
FIG. 14—2D Scanning Based 3D Display Concept using mirrors placed on top of polymer scanners illuminated by external laser diode sources
FIG. 15—In the display concept there is an array of 2D polymer/hybrid scanners in the horizontal axis of the display
FIG. 16—Vertical and Horizontal view of the display
FIG. 17—Back and Forth Movement of the Pitch-Size Modulatable Lenticular Screen
FIG. 18—Left and Right Movement of the Pitch-Size Modulatable Lenticular Screen
The light source 13, collimator 31, 2D micro display panels 60, and the lens system 32 in front of the micro display 60 in
Each 1D scanning module 101D creates a horizontal scan line by way of electromagnetic actuation in this preferred embodiment [9]. A magnet is placed onto the backside of the 1D scanning module 101D and modulated by an external electrocoil. In order to realize the screen 20 capable of emitting different color and intensity light to different directions from its pixels S, red, green, and blue LEDs, 13R, 13G and 13B are modulated individually during scan and the images for each color LED can be overlapped in space by introducing slight time-shifts in between R, G, B LED drive signals during the scan.
As illustrated in
The screen 20 is capable of diffusing light into a narrow angle in the horizontal direction and into a wide angle in the vertical direction—i.e., elliptically diffusing screen 20. A narrow angle is required in the horizontal direction as each screen pixel S on the display should emit light with different color and intensity to separate horizontal directions without any crosstalk between neighboring directions. The wide angle in the vertical direction is required as the display is designed to provide motion parallax only in the horizontal direction (i.e., the same image is received by the viewer at the same horizontal position and different vertical positions of the eye pupils.)
The number of different views for the display is the same with the number of independently controllable horizontal emission directions from the screen pixels S. In a preferred embodiment, there are 40 different views using 1° divergence for each emission direction and 40° scan angle. The resolution of the display can be calculated using the following relationship:
The number of voxels O (NT) fed into the data channel per frame in the 3D display system is given by the product of total number of LEDs and p:
NT=nhnvlp (3a)
Equivalently, number of voxels O (NT) can also be calculated using the total number of screen pixels S and ray directions:
NT=NhNvr (3b)
Table 1 provides an exemplary system design parameters for 2 million and 20 million voxels O with different display depths.
The table implies that the resolution of the system can be increased by increasing the number of 1D scanning modules 101D without altering the 1D scanning module 101D design or the screen 20 depth, resulting in a scalable architecture. Another implication of the table is that the screen 20 depth can be reduced by reducing p and increasing nh.
The optics for the system is rather simple and illustrated in
Plurality of microlens arrays 30M can also be used as the imaging lens 30 in front of each 1D scanning module 101D. There are different modes of microlens arrays that can be used to image the light sources 13 to the screen 20. The first mode is superposition mode as illustrated in
As can be seen in
The LED arrays will be driven with a LED driving IC 14 which will also be placed on top of the polymer 1D scanning platform 101D to produce a compact system with minimum electrical connections through the flexible members 11 of the 1D scanning module 101D that are connected to a fixed platform 12. The second way of LED driving will be using an external LED driving circuitry with a field programmable gate array (FPGA), complex programmable logic device (CPLD) or an ASIC. Placing the LED driving IC 14 on top of the 1D scanning platform 101D provides a more compact design and gives the opportunity of increasing the number of LEDs on a single FR4 scanner as fewer electrical signals 15 should be carried through the flexible members 11 of the 1D scanning module 101D. These signals 15 would be limited, in the case of an FPGA, with the FPGA supply voltages VCCO, VCCAUX and VCCINT, JTAG programming interface signals, 1 bit clock signal and 1 bit serial input data that would modulate the LEDS connected to the FPGA I/O pins. In this case, the number of the LEDs that can be driven will be limited with the number of I/O pins of the FPGA which can be quite high; more than four hundred with an I/O optimized FPGA as shown in
The LEDs are driven by pulse width modulation (PWM) method. N bit depth level PWM provides 2N different intensity levels. A counter is synthesized within FPGA whose output value is compared with a reference value for each single output pin and produces PWM LED drive signal. N-bit video input determines the LED drive pulse width.
The input video data frequency at which the data will be fed into the FPGA will be:
As an example, assume fD=60 Hz scan frequency—typical refresh rates of displays and l=30 (or 90 LEDs per module), dPWM=10-bit, n=1 (scanners controlled by each driver), p=100 pixels/LED (=200 modulations per cycle due to bidirectional scanning). In such a case, if 1 bit per color (dW=3) serial input video data is fed into the FPGA then 3.6 MHz clock frequency would be required. Taking into account the sinusoidal speed variation of the scanner during resonant operation, this average data rate need to vary by about a factor of 2 from the center to the edge of the scan line.
The whole display concept D3D is shown in
In the case of limited number of viewers, viewing the display from a limited field of view (FOV), the scanning angle of the scanners can be limited to a specific narrow angle with an offset angle enough to feed all the viewers in the limited FOV. Each actuated 1D scanning module 101D—electromagnetically in the above configuration—is applied a certain constant magnetic force according to the viewers' position in the FOV of the display. The 1D scanning modules 101D are scanned with an alternating magnetic force around this offset value to provide the left and right eye views simultaneously for the limited number of viewers. By this way, the display system D3D works more efficiently and the display will be brighter as the number of views is limited.
The above 3D display concept D3D can also be realized by using single laser diode or vertical cavity surface emitting laser (VCSEL) for each red, green and blue colors as the light source 13 of the display scanned with 2D scanning modules 102D instead of the 1D LED array for each red, green and blue colors scanned with 1D scanning modules 101D. Two different configurations can be designed for the system using 2D scanning. In the first configuration, the laser light sources 13R, 13G, 13B are placed on top of the 2D scanning modules 102D as shown in
The horizontal resolution calculation of the system is the same with the above 3D system. The only difference appears in the vertical resolution calculation. The vertical resolution is the number of the vertical screen pixels S addressed by each scanning module 102D. The optics for the system is simple, only an imaging lens 30 for each 2D scanning module 102D is required. The horizontal and the vertical cross section of the system can be seen in
Similar to the 1D scanning module 101D with 1D light source array, 2D scanning module 102D can also work with a constant force and actuate around a specific angle only to feed a limited number of viewers in a limited FOV. Similar to the 1D case, scanning in a narrower angle increases the efficiency of the system and brighter images the viewers receive.
A single viewer 3D display more appropriate for personal devices using scanning light concept can be realized by using a dynamic screen 40—e.g. an array of cylindrical lenses (lenticular sheet) in front of the light sources 13 as shown in
Filing Document | Filing Date | Country | Kind | 371c Date |
---|---|---|---|---|
PCT/IB08/01140 | 5/6/2008 | WO | 00 | 8/3/2010 |