BG66655B1 - Device and method for obtaining stereo images - Google Patents

Abstract

The devices and methods can be applied for obtaining three-dimensional relief images on an electronic surface - an electronic display or electronic paper. According to these, a 3D relief image is obtained on an electronic display or electronic paper (1) by generation of polygon patterns that are projected onto the flat surface of the screen (1), wherein to each point of the image infinite number of different values are assigned, with different deviation from the direction of the orthogonal polygon, which is projected onto the given point by varying the angle of view and the visible picture, thus displaying the shape of the 3D object by transmitting the main direction of the light reflected from the surface of the screen or the paper (1) by stereo pixels (13), which transmit different images for the different angles to the plane of the screen or the paper (1).

Description

Field of application

The present invention relates to devices and a method for producing stereo images, which are applicable for obtaining three-dimensional relief images on an electronic plane, an electronic screen or electronic paper.

BACKGROUND OF THE INVENTION

Methods and devices for obtaining a stereo image with the help of stereo pairs are known - two different images for the left and right eye. In 3D screens, these methods are realized using a stereoscopic (using glasses) and autostereoscopic method. Stereoscopic screens allow you to get a quality pseudo-stereo image - the image remains the same at different viewing angles. Stereoscopic screens do not depend on the number and position of observers, as the separation of the left and right images is done with the help of glasses. This simplifies and supports the technical realization of the stereo effect, but does not allow immediate visual perception without the use of glasses. Autostereoscopic screens are used to eliminate this shortcoming.

All known methods for obtaining autostereoscopic images (display based on parallax barrier, raster display, sequential temporary display) depend on the position of the observer and the location of his left and right eye.

In addition to the great technical difficulties associated with the separate translation of the image to the left and right eyes of each viewer, there is also a problem with the coincidence of the left image with the field of view of the right eye and vice versa. This happens when the viewers are too close or when their location changes abruptly. In addition, lenticular lens raster technology, used in both autostereoscopic and stereoscopic screens, creates a contradiction between convergence and accommodation. This causes unpleasant sensations and eye fatigue when perceiving the stereo image. These methods and devices also cannot transmit (show) the real parallax (the transformed image is actually a pseudostereo image).

Various mechanisms of visual perception of three-dimensional space are known. For example, the geometric perspective (gradient of magnitudes) and sfomato (chiaroscuro), the aerial perspective (reduction of sharpness and change of the total spectrum at a distance), used by artists in the Renaissance. The vertical perspective (height and position of the object in sight) used by artists in ancient China and Japan. Distribution of light and shadows on which the European school of painting and the three-dimensional illusion of Escher is based. Monocular parallax (parallax of movement or real parallax), vergent eye movements, accommodation of the eye crystal, the distribution of perceived objects in figures and backgrounds, the creation of barriers (closing surfaces), relative dimensions. But the most effective mechanism of volumetric vision is bilocular parallax (stereopsis), based on the differences (disparity) of monocular images perceived by the left and right eyes. The principle of operation of most mass-produced 3D screens is based on the perception of depth and remoteness with the help of stereopsis and accommodation. And for volumetric screens, the stereopsis serves to convey the shape of the 3D object. For the first time a complete image based on binocular parallax was created in 1885 by Charlemagne D. Almeida. His method, later called anaglyph, was based on the known distribution of the two-color image using light filters. In 1896, Bertie invented a method for obtaining autostereoscopic images (visible without the use of glasses) using a raster grid. And in 1908 Lipman invented the method of optical raster (penticular). With little change, these technologies have survived to this day. The principle of operation of stereoscopic 3D screens is based on the division of the stereo pair with the help of color light filters, polarization filters or optical shutters (locks), and the parallax-barrier method is used in most autostereoscopic screens.

Stereoscopic screens using red and blue (green) filters distort color information and cause a temporary disturbance of color perception. This disadvantage is avoided in screens using liquid crystal shutters (locks). To avoid image flicker, their switching time should be short enough. In Japan, such displays are used in the system of volume satellite TV. However, their application is limited, as the binocular parallax used to create a deep effect and different distances causes disturbances in the natural perception and leads to eye fatigue. Numerous parallax systems without glasses have been developed to eliminate the discomfort of using glasses. An implementation of the parallax barrier method is disclosed in U.S. Patent 5,969,850. The light from pixels forming subframes is transmitted through a vertical lumens in different directions, one direction for each viewpoint. The number of horizontal pixels on the 2D display limits the number of directions of the view. The contradiction between accommodation and convergence also persists. In addition, the display is very inefficient in terms of light. Only a small part of the light emitted by the pixels reaches the observer.

Another method and device for obtaining stereo images visible from different angles (angles) of the review are disclosed in patent WO 1998/034411. They use several 2D displays illuminated by laser LEDs to create a 3D image, and a special screen is used to transmit this image from different viewing angles. The number of 2D displays required to display a 3D image is comparable to the number of angles of view. A separate design system is used for each of the 2D displays, which leads to a very high cost.

A method is known for preparing flat focusing elements by means of micro-mirrors arranged in one plane and oriented at different angles (Patent US 6600591 B2, US 6870660 B2, US 7095548 Bl).

Such lenses, created on the basis of micro-mirrors, are used in many modern devices, video projectors, 2D and 3D TVs, etc. The micro-mirror lens is an example of a flat device that fully reproduces the front of the light wave reflected by a similar concave mirror element. Also known is a method for obtaining a 3D effect by means of micro-mirrors arranged in one plane and oriented at different angles of inclination (EP 0710508 / EP 1700640). The method and apparatus described in this patent make it possible to obtain simple 3D images such as an arc or hemisphere using permanent magnets. In this way, the stereo effect is achieved without the use of stereo pairs. The left and right eyes of the observer always see a different image, as the flat micro-mirror element completely reproduces the reflection of light from a similar volume element.

There are two types of 3D screens - screens that transmit real 3D images (volumetric displays), which are currently in the stages of experimental models and screens that transmit pseudo 3D images (stereoscopic and autostereoscopic screens), which are mass-produced by many manufacturers. Volumetric displays reproduce the volumetric image using voxels (volume pixeO-dots), located in a volume, not in a single plane, as in ordinary 2D screens. Examples of such displays are screens consisting of several liquid crystal screens creating a three-dimensional image at the expense of rapidly moving the corresponding image back and forth, or the so-called holographic display, which is a rapidly rotating screen on which a laser beam projects a three-dimensional image. , visible from all sides. Such displays create an image that transmits real parallax and do not cause any discomfort, as there is no contradiction between accommodation and convergence. But the technical difficulties arising during development and the high cost do not allow them to move from the development stage to serial production. In addition, the same transparency of the near voxel layers for the front layers, showing both the front and back of the 3D object, creates the unpleasant effect of image transparency. This requires the active development and implementation of pseudo-three-dimensional screens using technologies invented about a hundred years ago. (Independently by both Jacobson and Berthier around 1896, Frederick E. Ives of the U.S. in 1903, Estenave of France in 1906). In all these methods of obtaining a stereo effect, the principle of the stereo pair is applied (two different images for the left and right eye). Ie the binocular mechanism of stereoscopic perception characteristic of the human visual system is used. To create a stereo effect in this way, it is sufficient to simultaneously translate two images corresponding to different viewing angles, having different polarization, different spectral characteristics, or some other difference, allowing to separate the right and left image with the help of the eyes. The simplicity of this method makes it possible to obtain a quality pseudo-stereo image without serious deterioration of the contrast.

But the immediate perception of such a three-dimensional image is impossible. Attempts to obtain a stereo effect using a stereo pair without the use of glasses lead to technically complex and inefficient autostereoscopic systems. Instead of splitting the image using optical filters (glasses), these systems use the exact focusing of the stereo pair to the viewer's eyes. Since it is impossible for the viewer's eyes to be fixed in one position and different viewers may be in line with the screen, the problem of overlapping the left image in the field of view of the right eye and vice versa inevitably arises. The need to accurately monitor the position of the viewer's eyes makes the system extremely complex and expensive, and the contradiction between accommodation and convergence inherent in screens using lenticular technology leads to eye fatigue and makes watching (using) modern 3D screens insufficiently comfortable.

The methods that use binocular parallax to transmit the shape of 3D objects have avoided the disadvantages of systems that create a 3D effect by transmitting depth and distance. An example of the use of binocular parallax is the volumetric display with its inherent disadvantages.

Technical essence of the invention

It is an object of the present invention to provide a method and apparatus for producing stereo images in which an electronic screen or electronic paper composed of stereopixels emits or reflects light of different intensities from different viewing angles. In this way, each of the pixels transmits (reflects) not only the transition from light to dark, but changes its brightness depending on the angle of view (viewer's location), which is different for the left and right eye of the viewer (viewer). .

The method for producing a stereo image according to the invention comprises obtaining a 3D relief image on an electronic screen or electronic paper, in which polygon models are generated, which are projected on the flat surface of the screen, whereby infinitely many values of brightness (gradient) at different deviation from the direction of the orthotonal polygon, which is projected on a given point, whereby changing the angle of view and the visible image displays the shape of the 3D object by transmitting the main direction of the light reflected from the surface of the object on the screen or paper using stereo pixels that transmit a different image at different angles to the plane of the screen or paper. Preferably, the stereo pixels emit their own light or direct internal light rays.

The device for producing a stereo image according to the invention comprises an electronic screen or electronic paper. According to the invention, the screen or paper includes a liquid crystal layer for reproducing brightness and color in different parts of the 3D image, under which is located a micro-mirror layer for reproducing the polygon orientation of the 3D model, wherein each liquid crystal consists of optical domains. anisotropic material with a constant axis of symmetry for transmitting light in one direction only. The micro-mirror translucent layer transmits light transmitted from below in only one direction. According to another embodiment of the invention, the stereo imaging device comprises an electronic screen or electronic paper and control permanent magnets. The screen or paper consists of stereo pixels composed of movable microcapsules in which magnetic micro-mirrors are arranged, with control magnets arranged around each capsule to transmit the orientation of the 3D polygon object.

According to another embodiment of the invention, the stereo imaging device comprises an electronic screen or electronic paper, the screen or paper consisting of stereopixels, each of which is a liquid crystal pixel with a variable axis of optical symmetry of anisotropic optical material around which transparent electrodes are located for orientation of the 3D object.

According to the present invention, the use of stereopixels not only eliminates the disadvantages associated with the contradiction between accommodation and convergence, but also eliminates the defect of transparency - illumination of the reverse side inherent in volumetric screens. Furthermore, the method and devices according to the invention do not require the use of complex and expensive tracking and focusing systems. This allows you to get a better image quality and cheaper than volumetric and autostereoscopic screens.

Explanation of the attached figures

The present invention and most of its advantages will become more fully apparent from the following detailed description with reference to the accompanying drawings, in which:

Figure 1a shows a schematic transmission of a picture through the autostereoscopic screen;

Figure 1b shows a schematic representation of a picture by stereopixels according to the present invention;

Figure 2 shows a general diagram of stereopixels with micro-mirror layers and liquid crystal layers according to the present invention;

Figure 3a shows liquid crystal stereopixels with 8 domains of optical anisotropic material that transmit light in only one direction according to the present invention;

Figure 36 shows a general diagram of a stereopixel of anisotropic material according to the present invention, with a constant axis of symmetry comprising a plurality of domains;

Figure 4 shows a general diagram of a stereopixel according to the present invention, consisting of a movable microcapsule of transparent dielectric material, around which a system of control magnets is arranged;

Figure 5 shows a general diagram of a stereopixel according to the present invention, from a liquid crystal with a variable axis of symmetry;

Figure 6 schematically shows the orientation of the magnetic micro-mirrors when switching on different groups of control magnets.

An exemplary embodiment of the invention

The method of obtaining a stereo image involves obtaining a 3D relief image on an electronic screen or electronic paper 1 by generating polygon models that are projected on the flat surface of the screen 1, where at each point of the image are set infinitely many values of brightness (gradient ) at different deviations from the direction of the orthogonal polygon, which is projected on the given point, whereby changing the angle of view and respectively the visible image displays the shape of a 3D object by transmitting the main direction of the light reflected from the surface of the object on the screen; the paper 1 by stereopixels 13 which transmit a different image at different angles to the plane of the screen or the paper 1. The stereopixels 13 emit their own light or direct internal light rays.

The screen or paper 1, according to one embodiment of the embodiment, consists of stereo pixels 13, each of which includes a liquid crystal layer 2 for reproducing brightness and color in different parts of the 3D image, under which a micro-mirror layer 3 for reproducing the polygon orientation of the 3D model, in which each liquid crystal 5 consists of domains 6 of optical anisotropic material with a constant axis of symmetry for light transmission in one direction only. The micro-mirror translucent layer 3 transmits the light supplied from below in only one direction. According to another embodiment of the embodiment, the screen or paper 1 consists of stereopixels 13 composed of movable microcapsules 7 in which magnetic micro-mirrors 8 are arranged, wherein around each capsule are located the control permanent magnets 9 for transmitting the orientation of the 3D polygon. object.

It is possible for the screen or paper 1 to consist of stereo pixels 13, each of which is a liquid crystal pixel lie a variable axis of optical symmetry of anisotropic optical material around which transparent electrodes 10 are arranged to orient the 3D object.

As shown in FIG. 1a, the picture is transmitted through an autostereoscopic screen 1. The same stereo pair is transmitted in all directions, focusing on the left and right eyes of the viewer (viewer) by means of a complex electronic and mechanical system. According to FIG. 16, the image is transmitted through the stereo pixels 13, transmitting the shape of the 3D object through a micro-mirror layer in all directions. The multidomain LCD system is a system that transmits the shape of the 3D object in several selected directions. For both systems, the picture is transmitted in the corresponding given viewing angle.

According to FIG. 2, the lower layer 3 reproduces the polygon orientation of a 3D model through micro-mirrors. The top layer 2 reproduces the brightness and color in different parts of the 3D object. The micro-mirror layer 3 reproduces the shape of the sphere 4 and remains constant in all cases.

In FIG. Liquid crystal pixels 5 of 8 domains 6 are shown. Domains 6 consist of optically anisotropic materials that transmit light in only one direction.

In FIG. 36 shows the domains 6 made of optically anisotropic material that transmit light in only one direction. The multidomain stereopixel is an anisotropic material with a constant axis of symmetry (eg, a multidomain liquid crystal pixel).

In FIG. 4 shows a movable microcapsule 7 of transparent dielectric material and control magnets (eg permanent or electromagnets) -1, 2, 3, 4, 5, 6 and 7. The stereopixel is a movable microencapsulated part 7 of a magnetic micro-mirror that transmits the orientation to 3D polygon object.

In FIG. 5 shows the liquid crystal stereopixel an optically variable axis of symmetry of anisotropic material around which the transparent electrodes 10 are arranged.

In FIG. 6 shows the orientation of the magnetic micro-mirrors in different switching schemes of the control magnets, in which the orientation of the 3D polygon is transmitted: Picture 1: The magnets 1, 2, 4 and 6 are switched on; Picture 2: magnets 3 and 7 are on; Picture 3: magnets 2 and 6 are on; Picture 4: magnets 1 and 4 are on. According to the method, the shape of the 3D object is represented by transmitting the main direction of the light reflected from its surface through screens without an internal light source (stereo pixels are based on micro-mirrors (translucent mirrors for electronic paper)) or through screens with internally guided light rays (stereopixels are based on optical anisotropic filters). Reproducing the shape of the 3D object using stereo pairs reproduces the shape of the object from only one angle (angle). In this sense, the stereo image is just as static as the pseudo-relief image obtained using parallax occlusion mapping, multi-layer parallax mapping th. Such programs program the terrain with respect to one viewing angle and do not display a 3D image of the object at different viewing angles. Polygonal models are used to quickly reproduce a three-dimensional image visible from different viewing angles. If such a polygonal model is designed on a flat surface, then at each point of the image can be set not two values of brightness (left and right image), but an infinite number of values (gradient) at different deviations from the direction of the orthogonal polygon projected on the given point. In this way, when you change the viewing angle, the visible image will change and the stereo image will not be distorted when viewed from different distances. This allows this method to be used to create large screens, for example for use in cinemas, where it is impossible to limit the points (angles) of the view due to the considerable distance of the viewers. With a system using multidomain pixels that transmit the view of the subject at different angles from different angles of view, the stereo effect may decrease when viewed (viewed) from a great distance, as both viewer's eyes fall within the range of domains from one direction. In such a schematic, stereopixels have only one light-reflecting (or, in the case of the use of optical anisotropic filters - transmitted light) element. Ie each pixel 13 has one dominant direction of reflection (or emission, in the case of using an internal light source and optical anisotropic filters instead of micro-mirrors) of light. This is determined by the change in the position of the light-reflecting or light-emitting surface of the particular domain 6.

In all other directions, the intensity of the light radiation will decrease in direct proportion to the deviation from the main direction. In this way, by orienting the light-reflecting elements at a certain angle to the surface, the shape of the depicted object can be transmitted. If the depicted object is presented as a polygonal model and the light-reflecting elements are oriented at the same angles as the polygonal planes of the given model, then all sections of the virtual object presented on the flat surface will reflect light in the same direction as similar sections of the real object. . Because with a minimal change in the angle of view, the observed image will also change, the picture seen by the left and right eye will always be different. This provides three-dimensional perception (vision). The total brightness of the pixel and its color are determined by the upper liquid crystal layer. Ie the image consists of a 2D picture that is transmitted by the ordinary liquid crystal layer on the screen and a polygonal grid determining the orientation of the layer of micro-mirrors reflecting the light (bottom layer of the screen) of each participant in this image. For the implementation of the method in this case can be used movable micro-mirror elements or pixels based on micro-mirror magnetic fluids. Stereopixels based on micro-mirror liquids have a higher (faster) action than mechanically switchable micro-mirror matrices and have no limitation on the angle of rotation. In a polygonal three-dimensional model, the orthogonal direction relative to the surface of the depicted 3D object can be considered as the main vector of light distribution. Ie the direction perpendicular to the polygonal surface will correspond to the maximum reflection of light from the 3D object. The deviation from this direction is transmitted using the brightness gradient. By orienting the particles of the micro-mirror liquid perpendicular to this direction, the shape of the object can be recreated (the micro-mirrors are oriented parallel to the shape of the virtual object). In this case, the brightness and color of the object are transmitted not by the micro-mirror reflective layer, but by the upper liquid-crystal (LCD) layer. Ie the lighter (or brighter) part of the surface of the 3D object, visible from the side, is transmitted using the LCD (liquid crystal) layer, and the front surface will be seen as brighter using the lower micro-mirror layer. In this way, the image is formed by a 2D picture (geometric projection of the 3D image on the frontal plane) and the vector (direction) of the reflected light corresponding to each point (pixel) of the given 2D image. In addition to parameters such as brightness and color, each point (pixel) of such a 2D image will correspond to two values of the three-dimensional polar coordinate system: F-angle from 0 to 180 degrees horizontally and 0 angle from 0 to 180 degrees vertically. It is most convenient to use polygonal three-dimensional models to bring the image into this system. Then the orientation of the micro-mirrors will be parallel to the polygons of the polygonal model of the 3D image. Such an orientation of the micro-mirrors corresponds to the pixels of that part of the 2D image on which the projection of the given polygon is located. The orientation of the micro-mirrors of the individual stereopixels (on each one in particular) coincides with the angle to the front plane, under which the respective polygon is located. Ie The image consists of a 2D image that is transmitted by the ordinary liquid crystal layer on the screen and a polygonal grid that determines the orientation of the light directed at the micro-mirrors (bottom layer of the screen) of each part of the image. The polygonal model is based on the recording made from different points of view, or a pseudo-three-dimensional model generated by several (eight or more) flat images from such programs as 360 degrees of Freeedom, 3D photo builder professional, easypano modelweaver and others. Stereopixels based on micro-mirror fluid have a higher speed of action than stereo pixels based on a fully movable micro-mirror. Creating such stereopixels requires a magnetic fluid that combines high reflectivity with minimal reflection of diffuse scattered light. The magnetic micro-mirror liquid is obtained by dispersing very thin plates (foil) of a ferromagnetic or diamagnetic material coated with a metal reflective layer, or of a dielectric broadband mirror coating in an inert liquid. Magnetic-based micro-mirror particles can be prepared by any of the known methods for applying a thin film coating. The methods for layering thin strips (foil) are described in the "Encyclopedia of Industrial Chemistry of Ullmann" - fifth edition, p. 67, publishing house "Chemistry" Weinheim, Germany.

It is preferable to obtain thin foil strips of magnetic material (e.g. nickel, magnetite, hematite) coated with a light-reflecting layer (e.g. indium, silver, chromium, aluminum) by the PVD process. The micron-sized shredded foil strip is coated with a surfactant and dispersed in an organic liquid. To eliminate the diffuse scattering of light between the micro-mirror particles, the liquid may contain a UV-resistant organosoluble black colorant. To preserve the orientation of the micro-mirrors after switching (at rest), the liquid may contain from 0.5 to 4% thixotropic additive (such as modified polyamide VUK-430 or VUK-431).

Stereo pixels for screens using an internal (own) light source can be created using optical anisotropic materials such as liquid crystals, ulexite and others. The most popular (applicable) are liquid crystal based pixels. Their serious drawback (which can be eliminated by various means) is the limited viewing angle (of observation). This property, and in particular the presence of maximum transparency in the orthogonal direction (area of unobstructed view) and the sharp decrease in transparency when deviating from this direction, makes it possible to use pixels (based on liquid crystals) to change the degree of transparency in different angle to the screen plane. Multidomain systems such as MVA (Patent US 7474292, Patent US 7298353, Patent US 6992329) are known, which are used to increase the viewing angle. The multidomain system allows you to translate an image not only in the orthogonal direction, but also at an angle to the screen. With ordinary screens built on liquid crystals, this allows viewers to see the same image on the right, left, top and bottom. In our case, this allows the left and right eyes to see an image taken from a different angle (from different positions) - fig. 1.

Modern needs (to increase the viewing angle) are met by symmetric multidomain systems containing 8 (eight) domains in one pixel. Obviously, for a real 3D image (showing real parallax) these possibilities are insufficient. To display (translate) a pseudo-stereo image corresponding to the quality of 2D TVs, 6 (six) domains are enough horizontally, because when translating a pseudo-stereo image there is no decrease in contrast, but a smooth transition when deviating from the orthogonal direction of six from a different angle, is sufficient for a quality (clear) stereo effect at close range. All types of liquid crystal based pixels (in our case - domains) do not have a sharp limit of light transmission (see / do not see). Therefore, even in the presence of a total of four horizontal and four vertical viewing areas, the observer's eyes will not jump sharply from the left foreshortening area to the right and vice versa. When moving from one zone to another, the brightness of the pixel will be equal to the average value of the domains closest to the zone. When transmitting (displaying) real parallax, the 2D image of the subject will shift relative to the background, which will significantly reduce the contrast. Therefore, to transmit a real 3D image with high accuracy, it is advisable to use multidomain systems containing a larger number of domains in one pixel. For such a system (allowing the transmission of real parallax) in the reproduction of the image can be used multidomain pixels based on liquid crystals, transmitting (showing) several (16 or more) different images in 16 or more different directions. Accordingly, to translate such an image, there must be a recording made from different points of view or a pseudo 3D model created (generated) by several (eight or more) flat images by such programs as 360,

3D, 3D photo Builder professional, easypeano modelweaver, etc. Thus, there are two standards for stereo imaging:

1. By displaying real parallax via stereo pixels with 16 (sixteen) or more domains.

2. Without showing real parallax (as in modern 3D TVs) for systems with fewer domains in one pixel.

The orientation of the domains of each pixel in the system is fixed, as well as in ordinary multidomain pixels (MVA, etc.). The direction in which the light beam reflected by objects or the view from the virtual viewpoint was recorded, in the case of computer image processing, must correspond (coincide) with the direction of the light emitted by the stereopixel domains. Depending on the number of domains in the multidomain pixel, this angle will be different. Respectively, 8, 16 and more domain systems require their own image standards. Each pixel domain is responsible for a specific viewing angle. Horizontal domains can be more than vertical. If there are six domains horizontally, then each domain is responsible for a viewing angle of 30 degrees; if there are 12 domains horizontally, then each is responsible for a viewing angle of 15 degrees. In the first case, there will be only two domain zones vertically, in the second four. A symmetrical scheme 4/4, 8/8, etc. is also possible, but it is more inconvenient, as the eyes of the spectators are usually located in a horizontal direction. Accordingly, each pixel has 8 or 16 different degrees of brightness and color change in different viewing angles, which correspond to 8 or 16 different 2D images. This allows the eyes to see a different image from different points of view and when moving.

Claims (5)

Patent claims A method for producing a stereo image, comprising obtaining a 3D relief image on an electronic screen or electronic paper, characterized in that polygon models are generated which are projected on the flat surface of the screen, wherein at each point of the image are set infinitely many values of brightness at different deviations from the direction of the orthogonal polygon, which is projected on a given point, where by changing the angle of view and the visible image, the shape of the 3D object is displayed by transmitting the main direction of the reflected surface. the subject on the screen or paper (1) light through stereo pixels (13), which transmit a different image for different angles to the plane of the screen or paper (1). Method according to claim 1, characterized in that the stereopixels (13) emit their own light or direct internal light rays. Device for obtaining a stereo image, comprising an electronic screen or electronic paper, characterized in that the screen or paper (1) consists of stereo pixels (13), each of which includes a liquid crystal layer (2) for reproducing illumination and the color in different parts of the 3D image, under which a micro-mirror layer (3) is located to reproduce the polygon orientation of the 3D model, where each liquid crystal (5) consists of domains (6) of optical anisotropic material with a constant axis of symmetry for light transmission in one direction only. Device for obtaining a stereo image, comprising an electronic screen or electronic paper and control magnets, characterized in that the screen or paper (1) consists of stereo pixels (13) composed of movable microcapsules (7) in which they are located magnetic micro-mirrors (8), wherein the control magnets (9) are arranged around each capsule to transmit the orientation of the 3D polygon object. Device for obtaining a stereo image, comprising an electronic screen or electronic paper, characterized in that the screen or paper (1) consists of stereo pixels (13), each of which is a liquid crystal pixel (11) with a variable axis of optical symmetry of anisotropic optical material around which transparent electrodes (10) are located for orientation of the 3D object.