3D display system based on parallax modulation backlight

Abstract

The invention discloses a 3D display system based on parallax modulation backlight. The 3D display system comprises a backlight module, a 3D space box generation module and a front OLED transparent display screen. Wherein the 3D space box generation module is arranged between the backlight module and the front OLED transparent display screen, is composed of a microstructure imaging array and a parallax coding image carrier, and is used for carrying out parallax modulation on backlight and generating a virtual 3D space image with depth information, and the front OLED transparent display screen is used for displaying a conventional 2D image; the 2D image simulates three-dimensional depth clues such as spatial layering, perspective shielding, light and shadow gradual change and motion parallax through visual fusion with a virtual 3D space box, and enables an observer to generate vivid three-dimensional perception in combination with the Pepper's ghost principle.

Description

Technical Field

[0001] This invention relates to the field of display technology, and more specifically, to a 3D display system based on parallax modulation backlight, particularly an integrated display device that generates a virtual 3D space box through backlight parallax modulation and works in conjunction with a front-mounted liquid crystal display module to achieve 2D / 3D compatible display. Background Technology

[0002] With the rapid development of 3D display technology, various stereoscopic display devices have been widely used in advertising exhibitions, education and training, entertainment and interactive fields. Currently, stereoscopic display technology is mainly based on lenticular gratings. First, image encoding information from multiple viewpoints is generated on a 2D display screen. Then, the light emitted from the display panel is refracted using a lenticular grating array, allowing different parallax images to be imaged at different locations in space, thereby achieving a naked-eye 3D display effect.

[0003] However, in stereoscopic display technology using lenticular lens arrays, the sampling effect of the lenticular lens array will greatly reduce the resolution of the 2D display. For example, when using a 4K display to make an 8-viewpoint 3D display, the display resolution will drop to below 720p. In order to reduce the flicker and jump of the 3D display, a higher number of viewpoints is often required, which further exacerbates the resolution loss. This resolution loss is mainly determined by the following technical characteristics: (1) Optical structure of lenticular lens arrays: The lenticular lens unit refracts the light of each pixel to multiple directions, resulting in a reduction in the effective display area of ​​a single pixel; (2) The need for parallax separation: Multi-viewpoint display inevitably requires the pixel to be divided into multiple sub-pixels, and this division does not stop even in 2D mode, resulting in a decrease in pixel utilization; (3) Moiré patterns and crosstalk: The periodic structure of the lenticular lens array interferes with the black matrix of the liquid crystal panel, producing moiré patterns, which further reduces the image clarity.

[0004] Secondly, existing lenticular 3D displays face severe challenges in real-time disparity map generation. The real-time generation and display of multi-viewpoint disparity maps is a technical problem that is currently difficult to solve. Its essence lies in the imbalance between "multi-viewpoint data volume × dynamic scene complexity" and "limited computing resources." Specifically: at the data level, N viewpoints lead to an N-fold increase in data volume, exceeding the load capacity of conventional video processing frameworks; at the algorithm level, depth estimation, viewpoint interpolation, and crosstalk correction for dynamic scenes all require highly complex calculations, which are difficult to complete in milliseconds; at the hardware level, the computing power, bandwidth, and power consumption of existing GPUs / mobile chips cannot simultaneously meet the requirements of high viewpoint count, high resolution, and real-time performance. Therefore, lenticular 3D displays are currently mostly used for static images or low viewpoint count scenes, while real-time generation of high viewpoint count, high dynamic range video remains a technical bottleneck, requiring breakthroughs through lightweight algorithms, dedicated hardware acceleration, or cloud-based collaborative computing. These undoubtedly increase the complexity of display technology and manufacturing costs. Summary of the Invention

[0005] To address the aforementioned issues, this invention provides a 3D display system based on parallax modulation backlighting. It aims to achieve high-quality 3D display effects without sacrificing display resolution by integrating a lenticular grating with the backlight, rather than directly bonding it to the OLED transparent display screen. Furthermore, by utilizing 2D images combined with a virtual 3D space box to generate a 3D observation effect, it solves the problem of poor real-time parallax map generation in traditional solutions, allowing for realistic 3D perception simply by displaying a conventional 2D image.

[0006] To address the above technical problems, this invention proposes a 3D display system based on parallax modulation backlighting, comprising: Includes: a backlight module for providing an illumination source; a front-facing OLED transparent display screen for displaying two-dimensional images; and a 3D space box generation module disposed between the backlight module and the front-facing OLED transparent display screen; The 3D space box generation module includes a parallax coding map carrier and an optical microstructure array. The parallax coding map carrier is disposed between the backlight module and the optical microstructure array, and the optical microstructure array is disposed between the parallax coding map carrier and the front-facing OLED transparent display screen. The parallax coding map carrier is used to perform parallax modulation on the backlight, and after passing through the optical microstructure array, a virtual 3D space box with depth information is generated. The parallax coding map carrier is a transparent planar medium for printing a composite stereoscopic image, or a rear-facing OLED transparent display screen for displaying a composite stereoscopic image. The distance between the optical microstructure array and the parallax coding map carrier is set to be equal to the focal length of the optical microstructure array, so that the display plane of the parallax coding map carrier is located on the focal plane of the optical microstructure array. The optical microstructure array is a lens array composed of mutually parallel lenses of the same specification. The two-dimensional image displayed on the front-facing OLED transparent display screen is visually integrated with the virtual 3D space box, giving the observer a three-dimensional perception.

[0007] Furthermore, lenses include cylindrical lenses and Fresnel lenses.

[0008] Furthermore, the lens is a cylindrical lens, and the cylindrical lenses form a cylindrical lens grating unit.

[0009] Furthermore, the rear OLED transparent display screen and the front OLED transparent display screen are controlled independently. When the rear OLED transparent display screen is fully powered off and transparent, the 3D display system switches to the normal 2D display mode; when the rear OLED transparent display screen outputs a composite stereoscopic image, the 3D display system enters the 3D display mode.

[0010] Furthermore, the pixels of the synthetic stereoscopic image printed or displayed on the parallax-coded image carrier are composed of closely arranged square pixel units; each lenticular lens grating unit of the optical microstructure array covers the number of pixel units in the horizontal direction. The formula for calculating X is: X = 3 * (25.4 / LPI / cosα) / DOT In the displayed composite stereoscopic image, every three adjacent RGB subpixels constitute a pixel unit. In the formula, DOT is 1 / 3 of the distance between the centers of two adjacent pixel units on the composite stereoscopic image printed on the disparity-coded image carrier, or the distance between the centers of two adjacent subpixels on the displayed composite stereoscopic image; LPI is the actual number of lines per inch of lenticular lens; α is the tilt angle of the lens axis of the lenticular lens unit relative to the vertical axis of the disparity-coded image carrier.

[0011] Further, the synthesized stereoscopic image is synthesized according to the following steps: S1. Based on the required resolution H*V for synthesizing the stereo image, calculate N... tot Sampling is performed on images from each viewpoint; S2. Calculate the mapping table corresponding to the RGB components of the acquired viewpoint image with viewpoint number N and the RGB components of the synthesized stereo image. S3. N tot Each viewpoint image is cropped to match the outline of the carrier in the disparity-coded map and filled into the synthesized stereo image according to the mapping table.

[0012] Furthermore, in step S1, the vertical resolution of each viewpoint image is sampled as V / v, and the horizontal resolution is sampled as H / h, where v*h=N tot v takes the closest value Integer value.

[0013] Furthermore, the general formula for calculating the viewpoint number N in step S2 is: In the formula: mod represents the modulo operation; x and y are the horizontal and vertical coordinates of the sub-pixel of viewpoint number N in the synthesized stereo image; α is the tilt angle of the cylindrical lens axis relative to the vertical axis of the parallax coding map carrier; and P is the grating pitch of each cylindrical lens grating unit.

[0014] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention integrates a micro-optical array (such as a lenticular grating) with the backlight module and places it between the backlight and the liquid crystal panel, rather than directly attaching it to the front-facing OLED transparent display. In this way, the front-facing OLED transparent display can be used to display 2D images without sacrificing pixels due to parallax separation, thus preserving the full resolution of the original 2D image.

[0015] 2. Existing technologies require real-time generation of multi-viewpoint images, resulting in high computational load and latency, making it difficult to meet the demands of high dynamic range video. This invention constructs a fixed or variable "virtual 3D space box" as a depth stage. It only requires processing ordinary 2D images based on the lighting, perspective, and occlusion relationships within this spatial structure (e.g., utilizing the Pepper illusion principle) to trigger the human brain's stereoscopic perception. This eliminates the need to generate complex multi-viewpoint disparity maps, significantly reducing computational complexity and achieving low-cost, low-latency real-time conversion from 2D content to 3D experience. Attached Figure Description

[0016] Figure 1 A schematic diagram of the backlight 3D LCD display structure with light field modulation; Figure 2 This is a schematic diagram illustrating the effect of a virtual 3D space box. Figure 3 This is a schematic diagram illustrating the effect of combining a 2D image with a 3D spatial box. Figure 4 This is a diagram illustrating the occlusion relationship; Figure 5 This is a schematic diagram of motion parallax. Figure 6 A schematic diagram of the 3D space box generation module; Figure 7 This is a subpixel arrangement image; Figure 8 This is a schematic diagram of the grating tilt angle. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0018] Example 1 like Figure 1As shown, this embodiment provides a 3D liquid crystal display system based on parallax modulation backlight. Its overall structure, from back to front along the light propagation direction, includes: a backlight module 1, a 3D floating box generator 2, and a front-facing OLED transparent display screen 3. The three components are tightly bonded together by optical adhesive or a mechanical frame to form an integrated display panel.

[0019] The backlight module 1 is composed of a high-brightness LED array to provide a uniform and controllable surface light source.

[0020] The 3D space box generation module 2, located between the backlight module 1 and the front-facing OLED transparent display screen 3, generates a virtual 3D space box with a sense of depth, like a spatial stage, providing a spatial hierarchy for the display of 2D images. It consists of two parts: An optical microstructure array 5 (e.g., a lenticular grating or microlens array) is arranged close to the front-facing OLED transparent display 3, and a parallax coding map carrier 4 is arranged close to the backlight module 1, such as... Figure 6 As shown, the distance between the optical microstructure array 5 and the parallax coding map carrier 4 is set to be equal to the focal length of the optical microstructure array 5, so that the display plane of the parallax coding map carrier 4 is located on the focal plane of the optical microstructure array 5. The synthesized stereoscopic image is displayed on the parallax coding map carrier 4, which is a transparent planar medium, such as a static pattern with a specific spatial phase distribution, formed by printing high-precision transparent ink onto a PET film. The optical microstructure array 5 is a lens array composed of mutually parallel lenses of the same specification. The lenses include, but are not limited to, cylindrical lenses and Fresnel lenses. If cylindrical lenses are selected, then mutually parallel cylindrical lenses of the same specification form a cylindrical grating unit. In this embodiment, the synthesized stereoscopic image is a static, fixed pattern. Its design is based on a pre-calculated encoding of the target 3D virtual space structure (such as a floating cube, stage background, etc.), containing depth information encoding from multiple viewpoints. When the backlight module emits uniform white light that passes through this parallax encoding map, the light carries spatial modulation information. Subsequently, this modulated light undergoes directional refraction through a micro-optical array (such as a lenticular grating), reconstructing a virtual 3D space box with a sense of depth (i.e., a 3D Floating Box) in front of the observer. Figure 2 As shown.

[0021] The front-facing OLED transparent display 3 includes, from bottom to top, a transparent substrate, a first transparent electrode layer, an organic light-emitting functional layer, a second electrode layer, and an encapsulation layer. The OLED transparent display is used to display 2D image content (such as videos, text, and UI) required for user interaction. The transparent substrate uses glass or a flexible polymer as the substrate to achieve high transparency; the first transparent electrode layer uses indium tin oxide (ITO), which not only has good conductivity but also maintains transparency; the organic light-emitting functional layer emits light when current passes through it; the second electrode layer uses ITO material as the cathode; and the encapsulation layer is a transparent material. When current passes through the first transparent electrode layer, electrons and holes recombine in the organic light-emitting layer, exciting organic molecules to emit light, enabling the display of multiple color pixels such as red, green, and blue to form an image; since the first transparent electrode layer, the second electrode layer, and the transparent substrate are all transparent, light from non-emitting areas can penetrate the screen, thus achieving a transparent effect; when power is off, the OLED transparent display can achieve a transmittance of approximately 30%-40%, appearing transparent, allowing users to see the scene behind the screen.

[0022] During operation, the 3-pixel area of ​​the front-facing transparent OLED display can selectively block or allow light from behind to pass through, depending on the image content. When 2D image content is displayed, the corresponding pixel area is opaque, thus visually forming a foreground image; while the unblocked area allows background light from the 3D space box to pass through, maintaining the 3D virtual image structure of the background.

[0023] Specifically, this invention utilizes the Pepper's Ghost principle to perform real-time light and shadow processing on the 2D image presented by the front-facing OLED transparent display screen 3 according to the geometric structure of the 3D space box (e.g., Figure 3 (2) of the image, visually "embedded" into a virtual 3D space box generated by backlight modulation, and the final composite effect is as follows: Figure 3 As shown in (3) in the figure. The observer thus has a strong sense of stereoscopic depth.

[0024] The Pepper illusion achieves its realistic 3D illusion by simulating key three-dimensional depth cues through geometric optics and combining this with the perceptual mechanisms of the human visual system to guide the brain to "automatically complete" the three-dimensional structure. Its essence is not to reconstruct a real light field, but rather to trigger the inherent cognitive habits of humans regarding three-dimensional space through the interplay of light and shadow, perspective, and spatial superposition of planar images.

[0025] Therefore, this system simulates key depth cues in a 3D scene using the following methods: Light and shadow contrast: The 2D foreground image and the darker background of the virtual 3D space box create a strong contrast, suggesting that "the image is inside the 3D space box", which enhances the sense of spatial depth; Perspective consistency: 2D image content (such as people and objects) strictly follows the perspective rule of near objects appearing larger and distant objects appearing smaller (e.g., body proportions and size gradations), which conforms to the human eye's visual expectations of real three-dimensional space; Spatial occlusion: Foreground objects in a 2D image can "occlude" local areas of the 3D spatial box background (e.g., Figure 4 As shown in the figure, the depth judgment is triggered by the natural law of "near objects occluding distant objects"; Motion parallax: When an observer views the image from different angles (e.g., left view, front view, right view, see...) Figure 5 Due to the directional refraction characteristics of the micro-optical array, the background structure of the 3D space box will undergo relative displacement as the viewing angle changes; at the same time, although the 2D image is planar content, it is also perceived as an object with real depth due to its spatial binding relationship with the background and synchronous perspective adjustment, thus simulating the sense of movement of three-dimensional objects.

[0026] Thus, the observer sees a 2D foreground image embedded in a 3D virtual space constructed by backlighting, a parallax coding map, and a lenticular lens grating (see...). Figure 3 (3)). Since the 3D space box itself contains depth cues (such as perspective structure, light and dark gradation, occlusion relationship), and the 2D image is aligned with it in space and modulated by its light and shadow, the human eye and brain will automatically merge the two to produce a strong stereoscopic perception.

[0027] Example 2 In this embodiment, the parallax coding pattern carrier 4 in the 3D space box generation module 2 no longer uses a static printing medium, but instead uses a rear OLED transparent display screen, which is located between the backlight module 1 and the micro-optical array 5, and is used to generate variable parallax coding patterns in real time.

[0028] During operation, the rear OLED transparent display dynamically renders the corresponding synthetic stereoscopic image based on the current application scenario. This encoded image is essentially a set of precisely calculated spatial light intensity patterns, which, in conjunction with the micro-optical array 5, generate a virtual 3D stage with variable depth and structure in real time within space. Simultaneously, the front OLED transparent display 3 displays 2D content and, through a real-time rendering engine, performs matching lighting, perspective, and occlusion processing on the 2D image based on the geometry of the current 3D spacebox, ensuring that its visual performance is consistent with the lighting direction and spatial logic of the virtual stage (see [link to rendering]). Figure 3 (2)).

[0029] Since both the front and rear OLED transparent displays can be controlled independently, the system can achieve: ①3D / 2D combined display: 2D content is embedded as a foreground object into a dynamic 3D background; ②Mode switching without mechanical switching: When the parallax map output of the rear OLED transparent display is turned off (or set to full transparency), the system degenerates into a normal 2D transparent display; when the parallax map is enabled, it immediately enters 3D mode. It is worth noting that in this structure, the micro-optical array is located between the two front OLED transparent displays and the rear OLED transparent display, and is not directly attached to any display panel. Therefore, it does not divide the pixel structure of the front OLED transparent display, and fully preserves the resolution of the original 2D image (such as 4K, 8K), overcoming the problem of severe resolution degradation caused by pixel reuse in traditional cylindrical lens 3D displays.

[0030] To achieve accurate parallax modulation, this embodiment uses a sub-pixel mapping-based encoding method to obtain a synthesized stereoscopic image, which is then printed on the parallax encoding map carrier 4 of Embodiment 1, or displayed on the rear OLED transparent display screen of Embodiment 2.

[0031] like Figure 7 As shown, the height of each RGB subpixel of the synthesized stereoscopic image displayed on the disparity-coded map carrier 4 is three times its width. They are arranged on the disparity-coded map carrier 4 in the following order: each row is RGBRGB... until the entire disparity-coded map carrier 4 is filled, where every three adjacent RGB subpixels constitute a pixel unit. The synthesized stereoscopic image is composed of tightly packed pixel units. Each lenticular lens grating unit of the optical microstructure array 5 covers X RGB subpixels in the horizontal direction, and the number of pixel units is... The tilt angle of the cylindrical lens axis of the cylindrical lens grating unit relative to the vertical axis of the parallax coding image carrier 4 is α. The advantages of this setting are twofold: first, it can balance the resolution of the synthesized stereoscopic image in the horizontal and vertical directions, ensuring that its proportion is not distorted; second, it can reduce the moiré effect caused by the arrangement of LCD pixels.

[0032] X = 3 * (25.4 / LPI / cosα) / DOT, where DOT is the distance between the centers of two adjacent pixels in the composite stereoscopic image on the disparity-coded image carrier 4, and LPI is the actual line count of the lenticular lens per inch. The printed composite stereoscopic image is composed of closely arranged square pixel units.

[0033] The method for synthesizing stereoscopic images includes the following steps: S1. Based on the required resolution H*V for synthesizing the stereo image, calculate N... tot Sampling was performed on images from multiple viewpoints, and the 3D model was processed using software such as Blender and Unity according to a fixed angle range and a fixed angle difference (0.4° to 5°). totThe acquisition of viewpoint images involves setting the position and pose of the virtual camera based on the viewing angle range and angular difference, acquiring images of the 3D model at each viewpoint (i.e., position and pose), and rendering these images at each viewpoint to obtain the virtual image at that viewpoint, thus obtaining the disparity map. <N tot <100, where the virtual camera corresponding to one of the viewpoint images is positioned directly opposite the front-facing OLED transparent display screen, and this viewpoint image is a frontal view. Simulate image acquisition from different camera angles. The vertical resolution of each viewpoint image is sampled as V / v, and the horizontal resolution is sampled as H / h, where v*h=N tot v takes the closest value Integer values; the size of the synthesized stereo image is the same as the size of the disparity-coded image carrier 4; S2. Calculate the mapping table corresponding to the RGB components of the N acquired viewpoint images and the RGB components of the synthesized stereo image; S3. Collect N tot Each viewpoint image is cropped into a processed viewpoint image that matches the contour of the disparity-coded map carrier 4, and N is then processed. tot The processed viewpoint images are filled into the RGB components of the synthesized stereo image according to the mapping table.

[0034] S4. Transmit the synthesized stereoscopic image to the rear OLED transparent display screen or print it on the parallax coding map carrier 4.

[0035] The step S2 is based on Figure 8 The schematic diagram shown generates a mapping table, which indicates the viewpoint image number N corresponding to the sub-pixel (x, y) in the synthesized stereo image. The distance x between the sub-pixel (x, y) and the left edge of the cylindrical grating unit directly above it is... off x and y represent the horizontal and vertical coordinates of the synthesized stereo image. The coordinates are based on the coordinates (1, 1) of the top left sub-pixel of the synthesized stereo image. x increases by 1 to the right and y increases by 1 to the down.

[0036] Figure 8 The pitch of each cylindrical lens grating unit is P, the tilt angle of the lens axis relative to the vertical axis of the parallax coding map carrier 4 is α, and the width of the lens pitch P along the horizontal direction is P. x P x Represented as It is also equal to X; for x off For identical sub-pixels, their outgoing rays also refract at the same angle, thus originating from the same parallax image, corresponding to the offset x. off The viewpoint number N satisfies the following formula: Finally, the general formula for calculating the viewpoint number N is obtained. mod represents the modulo operation; Furthermore, since a pixel unit is generally composed of three sub-pixels: R, G, and B, therefore, we assume N R、 N G、 N B Let R, G, and B represent the viewpoint numbers corresponding to the three sub-pixels, respectively. Assuming the pixel at position (x, y) is located in row l (0 < l ≤ 4320) and column k (0 < k ≤ 7680), then: When N R、 N G、 N B When the value of N is not an integer, R、 N G、 N B Take the nearest integer value; when the calculated N... R、 N G、 N B When the value is 0, take N. R、 N G、 N B for .

[0037] In this way, each sub-pixel on the rear OLED transparent display or the parallax coding map carrier is assigned a viewpoint number, thereby forming a parallax coding map. When this coding map is combined with a micro-optical array, light from different viewpoints is directed in different directions, constructing a virtual 3D space box with depth layers in space.

[0038] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A 3D display system based on parallax modulation backlighting, characterized in that, include: The backlight module is used to provide a light source; Front-facing OLED transparent display for displaying two-dimensional images; And a 3D space box generation module disposed between the backlight module and the front OLED transparent display screen; The 3D space box generation module includes a parallax coding map carrier and an optical microstructure array. The parallax coding map carrier is disposed between the backlight module and the optical microstructure array, and the optical microstructure array is disposed between the parallax coding map carrier and the front-facing OLED transparent display screen. The parallax coding map carrier is used to perform parallax modulation on the backlight, and after passing through the optical microstructure array, a virtual 3D space box with depth information is generated. The parallax coding map carrier is a transparent planar medium for printing a composite stereoscopic image, or a rear-facing OLED transparent display screen for displaying a composite stereoscopic image. The distance between the optical microstructure array and the parallax coding map carrier is set to be equal to the focal length of the optical microstructure array, so that the display plane of the parallax coding map carrier is located on the focal plane of the optical microstructure array. The optical microstructure array is a lens array composed of mutually parallel lenses of the same specification. The two-dimensional image displayed on the front-facing OLED transparent display screen is visually integrated with the virtual 3D space box, giving the observer a three-dimensional perception.

2. The 3D display system according to claim 1, characterized in that, Lenses include cylindrical lenses and Fresnel lenses.

3. The 3D display system according to claim 1, characterized in that, The lens is a cylindrical lens, and the cylindrical lenses form a cylindrical grating unit.

4. The 3D display system according to claim 1, characterized in that, The rear OLED transparent display screen and the front OLED transparent display screen are controlled independently. When the rear OLED transparent display screen is fully powered off and transparent, the 3D display system switches to the normal 2D display mode. When the rear OLED transparent display screen outputs a composite stereoscopic image, the 3D display system enters the 3D display mode.

5. The 3D display system according to claim 1, characterized in that, The pixels of the synthetic stereoscopic image printed or displayed on the parallax-coded image carrier are composed of closely arranged square pixel units; each lenticular lens grating unit of the optical microstructure array covers the number of pixel units in the horizontal direction. The formula for calculating X is: X = 3 * (25.4 / LPI / cosα) / DOT In the displayed composite stereoscopic image, every three adjacent RGB subpixels constitute a pixel unit. In the formula, DOT is 1 / 3 of the distance between the centers of two adjacent pixel units on the composite stereoscopic image printed on the disparity-coded image carrier, or the distance between the centers of two adjacent subpixels on the displayed composite stereoscopic image; LPI is the actual number of lines per inch of lenticular lens; α is the tilt angle of the lens axis of the lenticular lens unit relative to the vertical axis of the disparity-coded image carrier.

6. The 3D display system according to claim 1, characterized in that, The synthesized stereoscopic image is synthesized according to the following steps: S1. Based on the required resolution H*V for synthesizing the stereo image, calculate N... tot Sampling is performed on images from each viewpoint; S2. Calculate the mapping table corresponding to the RGB components of the acquired viewpoint image with viewpoint number N and the RGB components of the synthesized stereo image. S3. N tot Each viewpoint image is cropped to match the outline of the carrier in the disparity-coded map and filled into the synthesized stereo image according to the mapping table.

7. The 3D display system according to claim 6, characterized in that, In step S1, the vertical resolution of each viewpoint image is sampled as V / v, and the horizontal resolution is sampled as H / h, where v*h=N tot v takes the closest value Integer value.

8. The 3D display system according to claim 7, characterized in that, The general formula for calculating the viewpoint number N in step S2 is: In the formula: mod represents the modulo operation; x and y are the horizontal and vertical coordinates of the sub-pixel of viewpoint number N in the synthesized stereo image; α is the tilt angle of the cylindrical lens axis relative to the vertical axis of the parallax coding map carrier; and P is the grating pitch of each cylindrical lens grating unit.

Images