Immersive naked-eye 3D visual impact display system with double-screen dynamic parallax enhancement

Abstract

The invention provides an immersive naked-eye 3D visual impact display system with enhanced double-screen dynamic parallax, and relates to the technical field of 3D display, and the system provides a wide-angle and highly-vivid 3D visual effect for a user without wearing auxiliary equipment through a unique double-screen architecture design in combination with an advanced dynamic parallax control technology. The core of the system is a dynamic parallax adjustment algorithm, the dynamic parallax adjustment algorithm can track the sight line and head position changes of a user in real time, 3D image parallax on a screen is intelligently adjusted, and it is ensured that the user can obtain consistent and stable stereoscopic visual experience at different angles and positions. Besides, the system adopts a high-precision 3D modeling and rendering technology and is combined with a transparent OLED display screen, so that the sense of reality and the visual impact of the 3D model are further enhanced. Through color matching and texture mapping algorithms, the system also realizes seamless fusion between the foreground and the background, so that the whole picture is more natural and harmonious.

Description

Technical Field

[0001] The present invention relates to the field of 3D display technology, and particularly to an immersive naked-eye 3D visual impact display system with enhanced dual-screen dynamic parallax. Background Art

[0002] In today's digital age, three-dimensional (3D) display technology has become a key area for enhancing the visual experience and is widely used in various industries such as entertainment, education, industrial design, medical, and scientific research. However, existing 3D display technologies still face many challenges and limitations in practical applications and are difficult to meet users' demands for high-quality, natural, and immersive visual experiences.

[0003] Traditional 3D display technologies are mainly divided into two categories: wearable 3D display devices and naked-eye 3D display technologies. Wearable 3D display devices, such as 3D glasses or head-mounted devices, although able to provide a certain stereoscopic visual effect, users often feel discomfort when wearing these devices for a long time, and even experience symptoms such as visual fatigue and dizziness. This discomfort seriously affects the user's viewing experience and limits the application of 3D technology in long-term use scenarios. In addition, wearable devices limit the user's freedom of movement to a certain extent. Users need to wear the device throughout the viewing process and cannot move and observe freely as in a natural environment. This limitation makes it impossible for users to obtain a natural and smooth visual experience when viewing 3D content, reducing the attractiveness of 3D technology.

[0004] On the other hand, traditional naked-eye 3D display technologies usually use multi-viewpoint or grating screen technologies to achieve a stereoscopic effect. However, these technologies generally have the problem of limited viewing angles. Users must be in a specific position directly in front of the screen, and even a slight deviation will cause image distortion or a weakened stereoscopic effect. This viewing angle limitation makes it difficult for naked-eye 3D technology to meet users' flexibility requirements in practical applications. At the same time, traditional naked-eye 3D technologies have deficiencies in parallax control and are difficult to achieve a strong and stable 3D effect with a deep sense. Especially in scenarios where 3D objects need to "rush out of the screen", existing technologies often cannot provide a high level of visual impact. This inaccurate parallax control not only affects the fidelity of the 3D effect but also reduces the user's immersion. In addition, existing naked-eye 3D display technologies have obvious deficiencies in the fidelity and immersion of the display effect. Due to the viewing angle limitation and inaccurate parallax control, users often cannot obtain a truly immersive experience when viewing. When dealing with complex scenes and dynamic content, existing technologies are also prone to problems such as image tearing and flickering, further affecting the user's visual experience.

[0005] In practical applications, these limitations have caused significant obstacles to the popularization and development of 3D display technology. In the entertainment field, in applications such as 3D movies and games, the existing technology cannot provide users with a natural and smooth visual experience. The discomfort and limited freedom of wearable devices make it difficult for users to maintain comfort during long-term viewing. The viewing angle limitation and inaccurate parallax control of traditional naked-eye 3D technology result in an insufficiently realistic 3D effect, failing to truly attract users. In the fields of education and industrial design, 3D display technology is widely used to display complex three-dimensional models and designs, but the limitations of the existing technology make it difficult for users to observe and understand these models. The viewing angle limitation and inaccurate parallax control lead to users being unable to accurately observe the details and spatial relationships of the models, affecting users' understanding and learning effects of the models.

[0006] In summary, the existing 3D display technology has many limitations in terms of user experience, viewing angle flexibility, parallax control accuracy, and display effect. These limitations not only affect the application effect of 3D technology in various fields but also restrict its further development and popularization. Therefore, there is an urgent need for a new technology that can overcome the shortcomings of the existing technology and provide a more natural, flexible, and realistic 3D visual experience. This new technology should be able to provide users with a wide viewing angle, accurate parallax control, and highly realistic 3D effects without the need to wear devices, thereby truly realizing an immersive visual experience.

[0007] In view of this, this application is proposed. Summary of the Invention

[0008] The present invention provides an immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement, which can at least partially improve the above problems.

[0009] To achieve the above object, the present invention adopts the following technical solutions:

[0010] An immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement, which includes: an environmental screen, a focus screen, a signal control component, a collection and monitoring module, and a moving module. The output end of the collection and monitoring module is electrically connected to the input end of the signal control component, the output end of the signal control component is electrically connected to the control end of the moving module, the output end of the moving module is electrically connected to the environmental screen and the focus screen, the focus screen is arranged behind the environmental screen, and the collection and monitoring module is configured to detect the viewing angle movement and position movement of the user;

[0011] Among them, the signal control component is configured to implement the following steps by executing the computer program stored in its internal memory:

[0012] Obtain the image data and reference materials to be projected, perform modeling processing on the image data and reference materials to obtain a presentation result, and import the presentation result into the focus screen and the environment screen, and adjust the color and lighting between the focus screen and the environment screen;

[0013] Obtain in real time the mobile monitoring results collected by the acquisition and monitoring module. When it is determined according to the mobile monitoring results that the user's perspective has changed, perform horizontal and vertical movement adjustments on the presentation result;

[0014] Judge whether there is area occlusion according to the mobile monitoring results. When it is judged that there is area occlusion, perform symmetric complement processing on the adjusted image to obtain a new depth map after symmetric complement;

[0015] Perform motion prediction and smooth transition preprocessing on the new depth map after symmetric complement, and separate the focus, foreground, and background based on the focus principle to end the adjustment process.

[0016] In summary, the immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement aims to solve problems existing in the existing 3D display technologies, such as viewing angle limitation, inaccurate parallax control, and poor user experience. Through the innovative dual-screen architecture design, combined with dynamic parallax control technology, high-precision 3D modeling and rendering technology, and advanced screen dynamic adjustment algorithms, this system provides users with an immersive 3D visual experience without the need to wear auxiliary devices.

[0017] Specifically, in terms of display technology, this system adopts the design concept of a main and a secondary dual-screen. The main screen is responsible for presenting the background image, creating a sense of depth and hierarchy; while the secondary screen uses transparent OLED technology to display the 3D models in the foreground, enabling the models to "jump out" of the screen and enhancing the visual impact. This way of combining the two screens not only improves the three-dimensionality of the image, but also ensures a stable and consistent 3D effect for users at different angles and positions through the dynamic parallax control algorithm.

[0018] To further improve the user experience, this system also introduces a variety of advanced algorithms. For example, the dynamic transition depth consistency algorithm can smoothly transition the image elements between the focus screen and the environment screen when the user's perspective changes, reducing visual fatigue. In addition, the algorithm based on the Gestalt principle can complement and optimize the 3D models to ensure that users can see the complete models from different perspectives and the animations are smooth and natural.

[0019] In terms of hardware configuration, this system has carefully selected a variety of high-performance devices, including high-resolution screens, high-quality audio systems, and advanced signal control hardware, etc., to ensure that the system can reach the best state in terms of display effects, sound quality performance, and stability. At the same time, the system is also equipped with devices such as depth cameras and stepper motors, which are used to monitor the user's position and head movements in real time, and control the dynamic adjustment of the screen through intelligent algorithms, further enhancing the user's interaction experience.

[0020] Generally speaking, through the innovative dual-screen architecture and dynamic parallax control technology, combined with advanced algorithms and high-performance hardware, this system brings an immersive 3D visual experience to users without the need to wear auxiliary devices. This system not only achieves a breakthrough in technology, but also brings a qualitative leap in user experience, with broad application prospects and important market value. Brief Description of the Drawings

[0021] Figure 1 is the front view of the immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement provided by the embodiment of the present invention;

[0022] Figure 2 is the side view of the immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement provided by the embodiment of the present invention;

[0023] Figure 3 is the rear view of the immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement provided by the embodiment of the present invention;

[0024] Figure 4 is the schematic diagram of 3D model parallax frame adjustment provided by the embodiment of the present invention. Detailed Description of the Embodiment

[0025] In order to make the purpose, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0026] Refer to Figures 1 to 4 As shown, the first embodiment of the present invention discloses an immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement, which includes: an environmental screen L1, a focus screen L2, a signal control component, a collection and monitoring module, and a moving module. The output end of the collection and monitoring module is electrically connected to the input end of the signal control component, the output end of the signal control component is electrically connected to the control end of the moving module, the output end of the moving module is electrically connected to the environmental screen L1 and the focus screen L2, the focus screen L2 is arranged behind the environmental screen L1, and the collection and monitoring module is configured to detect the viewing angle movement and position movement of the user;

[0027] Preferably, the signal control component includes an embedded GPU, a PCB main board, and a transmission module. The embedded GPU and the transmission module are configured on the PCB main board. The embedded GPU is an NVIDIA Jetson, and the transmission module is a fiber optic H·DMI or DisplayPort.

[0028] Preferably, the ambient screen L1 is a Micro-LED screen, the focus screen L2 is a transparent holographic display screen, the acquisition and monitoring module is a depth camera, and the moving module is a stepper motor.

[0029] Preferably, it further includes a vibration module and a sound module. The input ends of the vibration module and the sound module are electrically connected to the output end of the signal control component. Among them, the vibration module uses an electromagnetic vibration module and an aluminum alloy vibration diaphragm.

[0030] In this embodiment, the system provides an immersive 3D visual experience without the need for auxiliary devices for users through a unique dual-screen architecture and advanced dynamic parallax control technology. The main components of the system include an ambient screen L1, a focus screen L2, a signal control component, an acquisition and monitoring module, and a moving module. Among them, the output end of the acquisition and monitoring module is electrically connected to the input end of the signal control component, the output end of the signal control component is electrically connected to the control end of the moving module, and the output end of the moving module is electrically connected to the ambient screen L1 and the focus screen L2. The focus screen L2 is configured behind the ambient screen L1, and the acquisition and monitoring module is responsible for detecting the perspective movement and position movement of the user.

[0031] Specifically, the signal control component is the core control unit of the system. It includes an embedded GPU, a PCB main board, and a transmission module. These components are all configured on the PCB main board. The embedded GPU used is an NVIDIA Jetson, which is a high-performance graphics processing unit capable of quickly processing complex image rendering and parallax control algorithms. The transmission module uses a fiber optic HDMI or DisplayPort. This high-speed data transmission technology ensures the stable transmission of image signals, avoids data delay and packet loss phenomena, and thus ensures the smooth display of 3D images. Through this high-performance signal control component, the system can respond in real time to the perspective and position changes of the user, quickly adjust the parallax and focal length of the 3D image, and provide a stable and consistent 3D visual experience for the user.

[0032] The ambient screen L1 and the focus screen L2 are key components for the system to achieve 3D display effects. In this embodiment, the ambient screen L1 uses a Micro-LED screen, which has high resolution, high contrast, and excellent color performance, and can present a delicate and realistic background image to the user. The focus screen L2 uses a transparent holographic display screen, which can present a highly realistic 3D model, and due to its transparent characteristics, it can seamlessly integrate the foreground model with the background image, further enhancing the three-dimensional sense and depth of the image. Through this dual-screen architecture, the system can achieve a strong visual impact and immersive experience without the user having to wear auxiliary equipment.

[0033] In order to achieve accurate monitoring of the user's perspective and position, the acquisition monitoring module in this embodiment uses a depth camera. The depth camera can capture the user's head position and line of sight in real time, and its high-precision monitoring capability provides accurate data support for dynamic parallax control. When the user moves his head or changes the perspective during viewing, the depth camera can quickly capture these changes and transmit the data to the signal control component. Based on this data, the signal control component uses a dynamic parallax control algorithm to adjust the 3D model parallax and focal length on the focus screen L2 in real time to ensure that the user can obtain a stable and consistent 3D effect at different angles and positions. This real-time monitoring and adjustment mechanism not only improves the user's viewing experience, but also reduces visual fatigue caused by changes in perspective, allowing users to enjoy an immersive 3D visual feast for a long time.

[0034] The mobile module in this embodiment uses a stepper motor, and its main function is to dynamically adjust the distance between the focus screen L2 and the ambient screen L1 according to the user's perspective and position changes. Through this dynamic adjustment, the system can further optimize the 3D display effect and enhance the depth and layering of the picture. For example, when the user is close to the screen, the stepper motor drives the focus screen L2 to move closer to the ambient screen L1 to reduce the depth of field error; and when the user is away from the screen, the focus screen L2 moves away from the ambient screen L1 accordingly to enhance the depth of field effect. This dynamic adjustment mechanism not only enhances the three-dimensional sense of the picture, but also enables users to obtain the best 3D visual experience at different distances.

[0035] In addition to the above main components, this embodiment further includes a vibration module and a sound module. The input ends of the vibration module and the sound module are both electrically connected to the output end of the signal control component. The vibration module adopts an electromagnetic vibration module and an aluminum alloy vibration diaphragm, and this design can produce precise and controllable vibration effects, bringing a richer sensory experience to users. For example, when watching a 3D scene with strong visual impact, the vibration module can generate corresponding vibration feedback to enhance the user's immersion. The sound module adopts high-quality 3D surround sound equipment, and through precise sound positioning and surround effects, it further enhances the user's auditory experience. This multi-sensory experience combining vision and hearing makes the user seem to be in a real 3D scene, greatly enhancing the entertainment and attractiveness of the system.

[0036] The immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement provides users with an immersive 3D visual experience without wearing auxiliary devices through a unique dual-screen architecture, advanced dynamic parallax control technology, high-precision acquisition and monitoring module, and dynamically adjustable mobile module. The system not only achieves a breakthrough in technology but also brings a qualitative leap in user experience. Through a high-performance signal control component, high-quality display screens, precise monitoring devices, and enhanced modules for multi-sensory experience, it has achieved a technological leap in the field of naked-eye 3D display, with broad application prospects and important market value.

[0037] Among them, the signal control component is configured to implement the following steps by executing the computer program stored in its internal memory:

[0038] S1. Obtain the image data and reference materials to be projected, perform modeling processing on the image data and reference materials to obtain a rendering result, and import the rendering result into the focus screen and the environment screen, and adjust the color and illumination between the focus screen and the environment screen;

[0039] Specifically, step S1 includes: obtaining the image data to be projected, collecting reference materials, designing based on the image data and reference materials, and determining the shape, proportion, and dynamic characteristics of the image object;

[0040] Use the Blender tool to construct a low-polygon model of the shape, proportion, and dynamic characteristics of the image object, define the overall shape and skeleton layout of the object, and optimize its topological structure;

[0041] Use digital sculpting tools to perform detailed sculpting on the low-polygon model, add texture details and relief details of the object, generate high-resolution normal maps or displacement maps, and based on the UV settings, unfold the three-dimensional surface into a two-dimensional plane, and use Substance Painter to paint multi-channel texture maps for the object;

[0042] Add a skeleton system to the low - polygon model according to skeleton binding and weight settings, record the dynamic behavior based on motion capture, import the obtained data into the rendering engine, use PBR materials for real - time rendering or offline rendering to obtain the rendering result, and import the rendering result into post - production software for color correction, light effect enhancement, and sound effect synthesis to obtain the presentation result.

[0043] Import the 3D model of the presentation result into the focus screen and import the background of the presentation result into the environment screen;

[0044] Adjust the color and lighting between the focus screen and the environment screen using color matching and texture mapping algorithms to ensure that the background of the environment screen and the color change and lighting direction of the 3D model on the focus screen are consistent, as well as the consistency of color harmony and texture style.

[0045] In this embodiment, first, the system obtains the image data to be projected and relevant reference materials. These image data and reference materials (including hand - drawn sketches or simple 3D mock - ups) are the basis for modeling. By analyzing and processing them, the shape, proportion, and dynamic characteristics of the image object can be determined. This process is a key step in modeling, which provides an accurate design basis for the subsequent 3D model construction and ensures the authenticity and accuracy of the model. For example, when modeling a dinosaur model, by collecting reference materials on the shape design, proportion, and dynamic characteristics of the dinosaur, the form and actions of the dinosaur can be accurately restored, providing users with a realistic visual effect.

[0046] Next, use the Blender tool to construct a low - polygon model of the image object. This process includes defining the overall shape and skeleton layout of the object and optimizing its topological structure. The low - polygon model is the basis of 3D modeling, which provides a framework for subsequent detail sculpting and texture painting. By optimizing the topological structure, better performance and operability of the model in subsequent processing can be ensured. For example, when constructing a dinosaur model, the body structure and limb layout of the dinosaur can be initially defined through low - polygon modeling, laying a foundation for subsequent detail addition.

[0047] Then, use digital sculpting tools to perform detailed sculpting on the low-polygon model. This step includes adding texture details, undulation details, etc. of the object to generate high-resolution normal maps or displacement maps. Through digital sculpting, the surface details of the model can be made more abundant and realistic. At the same time, based on the UV settings, the three-dimensional surface is unfolded into a two-dimensional plane, and Substance Painter is used to paint multi-channel texture maps for the object. This step further enhances the visual effect of the model, enabling it to present more realistic light and shadow and texture when displayed. For example, in the detailed sculpting of a dinosaur model, by adding details such as skin texture and muscle undulation, the dinosaur model can look more vivid and real.

[0048] Add a bone system to the low-polygon model according to bone binding and weight settings, and record the dynamic behavior based on motion capture. This step enables the model to perform natural action displays and enhances the dynamic expressiveness of the model. Then arrange the scene environment, build a background for the dinosaur, such as an ancient jungle or a desert, and enhance the layering of the scene by adding detailed objects such as rocks, trees, and grass. At the same time, configure the lighting system, use the main light source to shape the three-dimensional sense of the dinosaur, use the auxiliary light to soften the shadows, use the backlight to enhance the contour effect, and enable global illumination and dynamic shadows to enhance the realism. After the model and the scene are prepared, import the obtained data into the rendering engine, and use PBR materials for real-time rendering or offline rendering to obtain the rendering result. PBR materials can provide more realistic material performance, enabling the model to present a realistic effect under different lighting conditions. Finally, import the rendering result into post-production software for color correction, light effect enhancement, and sound effect synthesis to obtain the final presentation result. This process further optimizes the visual and auditory effects and provides users with a more immersive experience.

[0049] Import the final presentation result into the system, where the 3D model is imported into the focus screen and the background is imported into the environment screen. The focus screen uses transparent OLED technology and can present a highly realistic 3D model, while the environment screen uses a high-resolution Micro-LED screen to provide an excellent background picture. Since the human eye tends to regard elements with similar shapes, colors, textures, etc. as a whole. In the dual-screen system, ensuring the similarity between the background and the foreground of the focus screen and the environment screen can increase the harmony of the overall picture and enable users to more naturally perceive the depth and hierarchy of the three-dimensional space.

[0050] Adjust the color and illumination between the focus screen and the ambient screen through color matching and texture mapping algorithms, ensuring that the color change and illumination direction of the background of the ambient screen are consistent with those of the 3D model on the focus screen. This is because color change and illumination play important roles in 3D images. When an object has a shadow, the human eye perceives a 3D effect; and the consistency of color and texture style makes the two screens visually seamless. This step is crucial for enhancing the overall sense and depth of the picture. For example, when the ambient screen shows a green forest background, the dinosaur model on the focus screen will have a moderate green tint, allowing the model to blend naturally with the background and enhancing the stereoscopic and immersive feeling of the picture.

[0051] Please refer to Figure 4 S2, obtain the mobile monitoring results collected by the acquisition and monitoring module in real time. When it is determined according to the mobile monitoring results that the user's perspective has changed, perform horizontal and vertical movement adjustments on the presentation result;

[0052] Specifically, step S2 includes: calculating the parallax amount Δd of the nth frame n , and its calculation formula is: where I p is the distance between the human eye and the 3D model, is the depth of the 3D model in the nth frame, is the distance from the user to the screen in the nth frame;

[0053] According to the parallax amount Δd of the nth frame n calculate the left-eye rendering position and the right-eye rendering position of the nth frame. The formula is: where is the original X coordinate of the nth frame of the 3D model;

[0054] Based on the left-eye rendering position and the right-eye rendering position, assuming that the user's head moves left and right and moves forward and backward collected by the acquisition and monitoring module, the parallax amount to be adjusted is: where k x and k z are both dynamic compensation coefficients;

[0055] Calculate the scaling ratio S of the 3D model after the nth adjustment n , S 0 = S is the original scaling ratio of the 3D model, α is the perspective deformation compensation coefficient, Δd0 is the parallax amount of the nth frame, and adjust the presentation result horizontally according to the scaling ratio and parallax amount of the 3D model.

[0056] In this embodiment, the acquisition monitoring module is used to obtain the user's perspective and position change information in real time. The acquisition monitoring module uses a high-precision depth camera, which can capture the user's head movement and the change of line of sight direction in real time. When the system detects a change in the user's perspective, it will trigger a dynamic adjustment mechanism for the rendering result. This process ensures that the user can obtain a stable and consistent 3D visual effect at any perspective, significantly improving the user's viewing experience.

[0057] Immediately afterwards, it enters the core parallax adjustment calculation stage. For each frame of image, the system first calculates the parallax amount of this frame to accurately reflect the influence of the user's perspective change on the parallax, providing the basic data for the subsequent rendering position adjustment. Based on the calculated parallax amount, the system further calculates the left-eye rendering position and the right-eye rendering position of the nth frame; thus presenting a 3D image with the correct parallax for the user. This precise parallax control not only enhances the realism of the 3D effect but also reduces the visual fatigue caused by perspective changes.

[0058] When the acquisition monitoring module detects the user's head moving left and right or moving forward and backward, the system needs to further adjust the parallax amount. These dynamic compensation coefficients can be optimized according to experimental data to ensure the smoothness and naturalness of the parallax adjustment. Through this dynamic compensation mechanism, the system can respond to the user's head movement in real time, further improving the stability and consistency of the 3D visual effect. Finally, the scaling ratio of the 3D model after the nth adjustment is calculated to ensure that the size and position of the 3D model remain consistent at different perspectives, avoiding model deformation or drift caused by perspective changes. This precise scaling adjustment mechanism further enhances the realism and immersion of the 3D effect.

[0059] Briefly speaking, when the user's perspective changes, the perspectives of the 3D model and the environmental screen background on the focus screen should be adjusted synchronously. At this time, the relative position, hue, brightness, etc. of the 3D model screen and the environmental screen background should transition smoothly to ensure that there is no obvious visual break between the two. This method uses the 3D model frame-by-frame segmentation method in computer graphics to ensure the consistency of image elements at different perspectives. For example, set 64 perspectives, and the person moves back and forth. The distance of this back-and-forth movement is 1 meter. Arrange the 64 perspectives within 1 meter, that is, when the person's eyes move back and forth, they can see the animation change of this 3D model.

[0060] The system incorporates an innovative algorithm, namely the dynamic transition depth consistency algorithm, which generates a smooth transition effect between the environmental screen and the focus screen as the position and perspective change. When the user changes the perspective, the 3D model on the focus screen and the background elements on the environmental screen transition through methods such as gradual change, blurring, and hue change, making the two appear as part of the same scene rather than disjointed images. The operation in step S2 brings a sense of comfort to the user. For example, when the 3D feeling of a perspective 3D image is relatively deep for us, and the depth of the next viewpoint after moving is relatively shallow, the pupils of the eyes will adjust during this process. The sudden change from deep to shallow will cause discomfort. Therefore, adding this dynamic transition depth consistency algorithm is to overcome and alleviate the comfort of the human eye.

[0061] Preferably, according to the dynamic adjustment formula D f = D0 + λ(Z u - Z0), calculate the dynamic adjustment distance between the focus screen and the environmental screen. D0 is the initial distance between the focus screen and the environmental screen in the default state, λ is the adjustment coefficient, Z u is the distance between the user and the screen, and Z0 is the reference user distance;

[0062] Among them, when it is judged that Z u > Z0, increase the dynamic adjustment distance D f between the focus screen and the environmental screen. When Z u < Z0, decrease the dynamic adjustment distance D f between the focus screen and the environmental screen;

[0063] On the basis of the dynamic adjustment distance D f between the focus screen and the environmental screen, add the user's moving acceleration and the user's moving speed to obtain the updated dynamic adjustment distance D f = D0 + λ(Z u - Z0)+ βv u + γa u , where βv u is the dynamic compensation according to the user's moving speed, γa u is the adjustment value optimized according to the user's acceleration, and both β and γ are weight coefficients;

[0064] According to the updated dynamic adjustment distance D f between the focus screen and the environmental screen, calculate the displacement s of the slide rail as s = kD f , and drive the moving module to drive the environmental screen and the focus screen to move according to the displacement of the slide rail, where k is the slide rail conversion factor.

[0065] In this embodiment, the horizontal movement solved in step S2 makes the 3D effect better, and then the visual effect is better during the vertical movement in step S3. The proximity principle of Gestalt holds that the closer the distance between objects, the easier it is for users to perceive them as a whole. Therefore, if the distance is too far, the 3D effect will decline. To solve this problem, through spatial sensing technology and linear rails, when the user approaches the screen, the distance between the focus screen and the ambient screen is appropriately shortened; when the user moves away, the distance between the focus screen and the ambient screen increases. In this way, the relative spatial relationship between the 3D model and the background always remains natural.

[0066] Specifically, the dynamic adjustment distance between the focus screen and the ambient screen is calculated according to the dynamic adjustment formula. In the default state, there is an initial distance between the focus screen and the ambient screen, which is the default setting of the system without user interaction. To adapt to the distances between different users and the screen, the system introduces an adjustment coefficient k and a reference user distance. Through these parameters, the system can dynamically adjust the distance between the focus screen and the ambient screen to ensure that users can obtain the best 3D visual effect at different positions.

[0067] Among them, when the user approaches the screen, that is, Z u < Z0, the system will reduce the distance between the focus screen and the ambient screen, thereby reducing the depth-of-field error and making the 3D effect more realistic. On the contrary, when the user moves away from the screen, that is, Z u > Z0, the system will increase the distance between the focus screen and the ambient screen to enhance the depth-of-field effect and further improve the 3D visual impact. This dynamic adjustment mechanism not only optimizes the 3D effect but also reduces visual fatigue caused by changes in the user's position, significantly improving the user experience.

[0068] To further optimize the adjustment process, the system adds the user's movement speed and acceleration to the dynamic adjustment distance between the focus screen and the ambient screen. By considering these dynamic factors, the system can respond more intelligently to the user's movement, thereby providing a smoother and more natural 3D visual experience. Among them, the weight coefficient is used to balance the influence of the user's movement speed and acceleration on the adjustment distance. By experimentally optimizing these weight coefficients, the system can ensure that when the user moves quickly or suddenly stops, the distance adjustment between the focus screen and the ambient screen remains smooth and natural, avoiding visual discomfort caused by too fast or too slow adjustment.

[0069] Finally, to achieve the physical movement of the focus screen, the system calculates the displacement of the slide rail based on the dynamically adjusted distance between the updated focus screen and the environment screen. The displacement of the slide rail is calculated through a slide rail conversion factor, which is used to convert the adjusted distance into the actual displacement of the slide rail. According to the calculated displacement of the slide rail, the system drives the moving module (such as a stepper motor) to drive the environment screen and the focus screen to move accordingly. This physical movement mechanism ensures that the distance between the focus screen and the environment screen can respond to the user's movement in real time, thereby providing the user with a continuous and consistent 3D visual experience.

[0070] In this step, a depth camera is used to identify the position of the user's head to estimate the distance. Algorithms such as YOLO, OpenCV, and MediaPipe are used for human detection. The distance is estimated by combining the FOV (field of view) of the camera and the face size. According to the estimated distance, a stepper motor + lead screw module is used to adjust the front and back positions of the screen.

[0071] S3. Determine whether there is an area occlusion according to the movement monitoring result. When it is determined that there is an area occlusion, perform symmetric completion processing on the adjusted image to obtain a new depth map after symmetric completion.

[0072] Specifically, step S3 includes: obtaining user perspective data (θ, φ) according to the movement monitoring result, and extracting the pixel depth information Z(x, y) of the 3D model, where θ is the user's current horizontal angle and φ is the user's current pitch angle;

[0073] Calculate the perspective projection matrix P(θ, φ) according to the user perspective data and the pixel depth information to simulate the 3D projection transformation under the user's perspective, and determine the area covered by the foreground object Z forcground (x, y) is the depth of the foreground object, Z background (x, y) is the depth of the background object, where when it is determined that the depth Z forcground (x, y) of the foreground object is less than the depth Z background (x, y) of the background object, this point is occluded;

[0074] Calculate the depth gradient of adjacent pixels on the screen And filter out the high-frequency regions To determine the occlusion boundary, where, is the partial derivative of Z with respect to x in the horizontal direction, representing the depth change rate of adjacent pixels in the x direction, is the partial derivative of Z with respect to y in the horizontal direction, representing the depth change rate of adjacent pixels in the y direction, and T is the occlusion boundary threshold;

[0075] According to the curve interpolation algorithm, the edges of the occluded area are geometrically contoured to obtain an estimated shape of the occluded area, and its mathematical expression is: P(t) = (1 - t) 3 P0 + 3(1 - t) 2 tP1 + 3(1 - t)t 2 P2 + t 3 P3, and the Laplace interpolation is used to extend the edges of the occluded area, and its mathematical expression is: where P0, P1, P2, and P3 are all known boundary points, t is the time parameter, and its value ranges from 0 to 1. ncighbor is the set of 4-neighbor pixels of the pixel (x, y)(x, y)(x, y), that is, the pixels above, below, left, and right (which can be expressed as: (x + 1, y), (x - 1, y)

[0076] , (x, y + 1), (x, y - 1), Z(x i , y i ) is the known depth value of the neighbor pixel (x i , y i ) for assisting in interpolating the depth of the current pixel, and Z(x, y) is the depth value at the screen pixel coordinates (x, y);

[0077] Calculate the symmetry center C(x, y) = ∑Z(x + i, y + j)Z(x - i, y - j) of the image, find the coordinate of the best symmetry axis position C(x, y), and according to the formula Z ′ (x, y) = Z(2C x -x, 2C y -y) to complete the symmetry center, and obtain a new depth map after symmetry completion. Among them, the depth value of its symmetric point coordinates 2C x -x, 2C y -y is used for filling, and Z ′ (x, y) is the depth value after completion.

[0078] In this embodiment, both horizontal and vertical problems are solved, but there will also be occlusions in all areas due to a certain visual change, which is a flaw. Next, the flaw needs to be repaired through an algorithm. The closure principle of Gestalt indicates that the human eye will automatically fill in the missing information to form a complete image. In the 3D model display, if part of the model is occluded due to a change in the viewing angle, the system can use the closure principle to complete it, so that the user can perceive a complete model. Based on the 3D model completion algorithm of the Gestalt closure principle, among them, three modules are implemented by algorithms. The first module is depth and geometric reasoning, and the second module is contour inference and interpolation completion, which is implemented through code.

[0079] Specifically, it is determined whether there is regional occlusion based on the mobile monitoring results obtained by the acquisition and monitoring module. The mobile monitoring results include changes in the user's perspective, and this data is crucial for determining which parts of the 3D model may be occluded. When the system determines that there is regional occlusion, it triggers a process of symmetric completion for the adjusted image. This process can not only repair the occluded area, but also enhance the overall sense and realism of the 3D model, enabling the user to see the complete model from any perspective.

[0080] First, the real-time angle information of the user is obtained through a depth sensor or a head-tracking camera, that is, the user's perspective data is obtained, including the user's current horizontal angle and pitch angle, and at the same time, the pixel depth information of the 3D model is extracted. This data is the basis for subsequent occlusion detection and completion processing. Immediately afterwards, it is calculated which parts of the model are occluded by foreground objects under the user's current perspective. Through the user's perspective data and pixel depth information, the system calculates the perspective projection matrix to simulate the 3D projection transformation under the user's perspective. This process can help the system determine which areas are covered by foreground objects. Specifically, the system compares the depth of the foreground object and the depth of the background object. When the depth of the foreground object is less than the depth of the background object, that is, the depth of the foreground object < the depth of the background object, this point is judged to be occluded. This accurate occlusion detection method ensures that the system can accurately identify the areas that need to be completed, providing reliable data support for subsequent completion processing.

[0081] Subsequently, the depth gradient of adjacent pixels on the screen is calculated to filter out high-frequency regions (regions with large gradient changes), thereby determining the occlusion boundary and filtering out regions where the depth gradient is greater than the occlusion boundary threshold. These high-frequency regions usually correspond to boundary positions with large depth changes and are important signs of the occlusion boundary. By accurately calculating and filtering the occlusion boundary, the system can provide clear boundary information for subsequent completion processing, ensuring that the completed image is more natural and coherent visually.

[0082] After determining the occlusion boundary, the system uses a curve interpolation algorithm to perform geometric contour completion processing on the edge of the occluded area. This process provides a basis for subsequent completion operations by estimating the shape of the occluded area. Through this curve interpolation method, the system can generate a smooth geometric contour, making the completed area naturally blend with the surrounding environment in terms of shape. In addition, the system also uses Laplace interpolation to expand the edge of the occluded area. Through Laplace interpolation, the system can effectively expand the edge of the occluded area, further enhancing the coherence and naturalness of the completed area. This interpolation method can not only fill the occluded area, but also make the completed image consistent with the surrounding environment in terms of depth, thereby reducing the visual abruptness.

[0083] Finally, calculate the symmetry center of the image, find the coordinate of the optimal axis of symmetry position, so as to determine the axis of symmetry position of the model, and provide a reference for symmetry completion. After determining the symmetry center, the system completes the symmetry center according to the formula to obtain a new depth map after symmetry completion. The principle is that if the object is symmetric (such as a sphere, a cylinder), then the other side of it can be used as a reference to complete the occluded part. Through symmetry completion, the system can use the symmetry of the model to fill in the occluded area, making the 3D model more complete and realistic visually. This symmetry completion method can not only repair the occluded area, but also enhance the overall sense and realism of the model, enabling users to see the complete 3D model from any perspective.

[0084] Briefly speaking, in step S3, first, use depth and geometric reasoning. The first step: perspective analysis and occlusion detection. The second step: input data: obtain the user's perspective and the depth map (Z-buffer) of the 3D model on the screen. The third step: occlusion detection: calculate which parts of the model are occluded by foreground objects at the current user angle. The fourth step: depth difference calculation: detect the position of depth mutation of the model on the screen (judge which areas are occlusion boundaries). Secondly, use contour inference and interpolation completion. The first step: boundary interpolation: establish geometric contours at the edges of the occluded areas, and fill in the blanks using curve interpolation or boundary extension methods. The second step: symmetry completion: for regular structures (such as spheres, cylinders, etc.), use symmetry and geometric characteristics to speculate on the occluded parts. The third step: multi-frame fusion: if the user's perspective shifts within a short period of time, refer to the information of the previous and next frames to speculate on the missing areas. The fourth step: finally, optimize the interpolation completion algorithm: through deep learning algorithms, the system can not only complete according to the known model data, but also automatically predict and generate a complete dynamic model according to physical motion laws (such as joint, muscle, and bone structures, etc.). When some areas are invisible due to perspective changes, the system will generate supplementary data according to the continuity of the model and physical principles.

[0085] S4. Perform motion prediction and smooth transition preprocessing on the new depth map after symmetry completion, and separate the focus, foreground, and background based on the focus principle to end the adjustment process.

[0086] Specifically, step S4 includes: predicting the future perspective of the fish guard according to the Kalman prediction algorithm and motion interpolation extreme. Among them, is the predicted value of the user's horizontal angle at the next moment, is the predicted value of the user's pitch angle at the next moment, v t is the angular velocity of the user's perspective, a t is the perspective acceleration, ω t is the pitch angle change rate, and Δt is the frame interval time.

[0087] Calculate the future positions of the key points of the 3D model according to the Kalman prediction algorithm. The formula is: where X t is the current frame position of the 3D model, V t is the model velocity vector, A t is the model acceleration, R t is the current rotation matrix of the model, Ω t is the model angular velocity, is the predicted value of the position at the next moment, is the predicted value of the rotation angle at the next moment;

[0088] Use the motion interpolation algorithm to create smooth transition frames between the key frames of the 3D model. Use cubic Bézier interpolation to calculate the intermediate frames. The formula is: P(t) = (1 - t) 3 P0 + 3(1 - t) 2 tP1 + 3(1 - t)t 2 P2 + t 3 P3, where P0 and P3 are the positions of the front and back key frames, and P1 and P2 are the control points;

[0089] According to the formula seamlessly connect the perspectives of the 3D model so that the model always maintains smooth motion under different perspectives. Among them, X left , X right are the position data of the model in two adjacent perspectives, and u is the interpolation coefficient;

[0090] Based on the focus principle, use image processing algorithms to adjust the dynamic brightness and contrast of the 3D model on the focus screen to separate the focus, foreground, and background.

[0091] In this embodiment, the above steps have completed the adjustment of horizontal, vertical, and filling in missing information. Since it is an animated display of the 3D effect, it is necessary to ensure that the animation is very smooth, as any lack of smoothness will affect the effect. This is because the movement or arrangement of objects will be presented along a continuous path. In 3D animation, the principle of continuity can ensure the natural smoothness of the model's movement and avoid motion breaks or incoherence caused by changes in the viewing angle. Therefore, through the smooth transition of skeletal animation, based on skeletal animation technology, the motion data of the 3D model is synchronized with the viewing angle change to ensure that the model's movement remains consistent and coherent regardless of the viewing angle from which the user observes. For example, when the dinosaur model is walking, the system will dynamically adjust its pace according to the position and viewing angle of the user's head, making it appear smooth and natural. It also includes motion prediction and interpolation algorithms, using interpolation algorithms to predict and smoothly transition the 3D model animation under the user's viewing angle. Whether the user is quickly rotating the viewing angle or moving smoothly, the system ensures the natural continuation of each frame of the animation by calculating the transition between key frames. Seamless viewing angle connection: When the user quickly changes the viewing angle, the system avoids incoherence in motion and ensures that the actions of the model (such as running, jumping, attacking, etc.) can smoothly transition under different viewing angles. For example, the pace of the dinosaur when running will be automatically adjusted according to the user's viewing angle, without any interruption or misalignment of the actions.

[0092] When the user quickly rotates the viewing angle, it is necessary to predict the motion state of the model in the next few frames. When moving smoothly, it is necessary to ensure the naturalness of the key frame motion and avoid stuttering or frame skipping. Specifically, the Kalman prediction algorithm and motion interpolation algorithm are used to predict the user's future viewing angle. This process is based on the user's current viewing angle data, including horizontal angle, pitch angle, viewing angle angular velocity, viewing angle acceleration, and pitch angle change rate, etc. Through these data, the system can accurately predict the viewing angle position of the user at the next moment. This prediction mechanism can adjust the display effect of the 3D model in advance to ensure that the user can still obtain a smooth visual experience when quickly moving the head, significantly reducing the visual delay and stuttering phenomena caused by viewing angle changes.

[0093] Immediately afterwards, the 3D motion prediction system needs to ensure that the actions of the 3D model do not mutate or misalign due to viewing angle switching, so it predicts the object position, velocity, and angular momentum. Calculate the future positions of the key points of the 3D model according to the Kalman prediction algorithm. This process takes into account the dynamic parameters of the current frame position, velocity vector, acceleration, rotation matrix, and angular velocity of the 3D model. Through these parameters, the system can accurately predict the position and rotation angle of the model's key points at the next moment. This precise motion prediction can not only ensure the accurate position and posture of the 3D model under different viewing angles, but also further enhance the realism and immersion of the visual effect.

[0094] To further optimize the motion effect of the 3D model and avoid sudden action changes, the system uses a motion interpolation algorithm to create smooth transition frames between key frames. This process adopts the cubic Bézier interpolation algorithm. By calculating the positions of the intermediate frames, it ensures that the motion of the model is smooth and natural from different perspectives. This is to ensure that actions such as a dinosaur running, jumping, and attacking do not show misalignment when the perspective changes. Through this interpolation algorithm, the system can generate smooth transition frames, making the motion of the 3D model more fluent and avoiding sudden motion changes and lags caused by perspective changes. This is to ensure that when the perspective changes, the transition of the model between different perspectives is more natural and does not jump abruptly.

[0095] In addition, the system also performs seamless perspective connection processing on the 3D model to ensure that the model always maintains smooth motion from different perspectives. This process realizes this seamless perspective connection processing through the interpolation algorithm. The system can ensure the motion coherence of the 3D model from different perspectives, further enhancing the user's immersive experience.

[0096] Finally, the system separates the focus, foreground, and background based on the focus principle; among them, the focus principle helps the human eye distinguish the relationship between the foreground and the background. The 3D model on the focus screen should always become the visual focus, while the environment screen acts as the background. By dynamically adjusting means such as parallax, brightness, and contrast, a clear visual separation between the two can be ensured; this step is also to enhance the depth effect. This process is realized through an image processing algorithm. By dynamically adjusting the brightness and contrast of the 3D model on the focus screen, the foreground model is made more prominent, while the brightness and contrast of the background are reduced, thus realizing the separation of the focus, foreground, and background. This separation mechanism can not only enhance the visual effect of the 3D model but also further improve the user's visual comfort, enabling the user to focus on the foreground model more naturally during the viewing process without being disturbed by the background.

[0097] Specifically, first, ensure that the focus model is prominent. When the user gazes at the focus screen, the system will automatically adjust the brightness, contrast, and edge highlights of the focus screen model to make it more prominent. At the same time, the brightness and contrast of the background will be correspondingly reduced to ensure that the foreground model becomes the focus visually. Secondly, perform dynamic brightness and contrast adjustment. Use the image processing algorithm to dynamically adjust the display effect of the 3D model on the focus screen, enhancing its contrast and color saturation, so that the foreground and the background are visually separated. For example, when the background is a dark-toned forest, the color and brightness of the 3D model will be adjusted to a more vivid tone to highlight the shape and actions of the model.

[0098] In summary, the immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement aims to solve the problems existing in existing 3D display technologies, such as viewing angle limitations, inaccurate parallax control, and poor user experience, through innovative technical means. The system provides an immersive 3D visual experience that users can enjoy without wearing auxiliary devices through a unique dual-screen architecture, dynamic parallax control, high-precision 3D modeling and rendering technologies, and advanced screen dynamic adjustment algorithms.

[0099] In terms of the dual-screen architecture, the system adopts a main and secondary dual-screen design. The main screen displays the depth background, and the secondary screen uses transparent OLED technology to display the foreground 3D model. This design not only enhances the sense of hierarchy of the scene but also uses the dynamic parallax control algorithm to capture the viewer's line of sight direction and head position in real time, intelligently adjusting the parallax and focal length to ensure that users can obtain a stable and consistent stereoscopic visual effect at different viewing angles. This dynamic adjustment mechanism significantly improves the realism and immersion of the 3D visual effect, reduces visual fatigue caused by viewing angle changes, and enables users to enjoy an immersive 3D visual feast for a long time.

[0100] In terms of 3D modeling and rendering, the system uses Blender modeling and rendering technologies to build high-precision 3D models, ensuring the realism and stereoscopic depth of the models on the transparent OLED screen. Through color matching and texture mapping algorithms, the system achieves seamless fusion between the foreground model and the background, further enhancing the visual impact. In addition, the system also introduces various algorithms based on the Gestalt principle, such as the closure principle algorithm and the continuity principle algorithm, to complete and optimize the 3D models, ensuring that users can see the complete models at different viewing angles and that the animations are smooth and natural. The application of these technologies not only improves the visual effect of the 3D models but also enhances the user's visual comfort and immersion.

[0101] In terms of dynamic parallax control, the system uses a high-precision acquisition and monitoring module to obtain information on the user's viewing angle and position changes in real time and dynamically adjusts the parallax and scaling ratio of the 3D image according to this information. This real-time adjustment mechanism not only ensures the stability and consistency of the 3D effect but also reduces visual fatigue caused by user movement. In addition, the system also introduces a dynamic transition depth consistency algorithm, which can smoothly transition the image elements between the focus screen and the environmental screen when the user's viewing angle changes, further enhancing the user's visual experience.

[0102] In terms of hardware configuration, the system has carefully selected a variety of high-performance devices, including high-resolution Micro-LED screens, transparent holographic displays, high-quality 3D surround sound devices, electromagnetic vibration modules, and advanced signal control hardware. The combination of these high-performance devices not only improves the display effect and sound quality performance of the system, but also enhances the user's multi-sensory experience. For example, the vibration module can generate precise and controllable vibration effects, bringing a more rich sensory experience to users; the high-quality audio module further enhances the user's auditory experience through 3D surround sound effects. The enhancement of this multi-sensory experience makes users seem to be in a real 3D scene, greatly improving the entertainment and attractiveness of the system.

[0103] Briefly speaking, the immersive naked-eye 3D visual impact display system with dual-screen dynamic parallax enhancement provides users with an immersive 3D visual experience without wearing auxiliary devices through an innovative dual-screen architecture, dynamic parallax control technology, high-precision 3D modeling and rendering technology, and advanced screen dynamic adjustment algorithms. The system has not only achieved a breakthrough in technology, but also brought a qualitative leap in user experience. With high-performance hardware configuration and advanced algorithm support, the present invention has achieved a technological leap in the field of naked-eye 3D display, and has broad application prospects and important market value.

[0104] The above is the preferred implementation manner of the present invention. It should be noted that for those of ordinary skill in the art of this technology, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A dual-screen dynamic parallax-enhanced immersive naked-eye 3D visual impact display system, characterized in that: include: An environment screen, a focus screen, a signal control component, a collection monitoring module, and a mobile module, wherein the output end of the collection monitoring module is electrically connected to the input end of the signal control component, the output end of the signal control component is electrically connected to the control end of the mobile module, the output end of the mobile module is electrically connected to the environment screen and the focus screen, the focus screen is arranged behind the environment screen, and the collection monitoring module is configured to detect the user's viewing angle movement and position movement; The signal control component is configured to implement the following steps by executing a computer program stored therein: Acquire image data and reference materials to be projected, perform modeling processing on the image data and reference materials to obtain a presentation result, and import the presentation result into a focus screen and an environment screen, and adjust the color and lighting between the focus screen and the environment screen; Acquire the motion monitoring results collected by the collection and monitoring module in real time, and when it is determined according to the motion monitoring results that the user's viewing angle has changed, adjust the presentation results in the horizontal and vertical directions; Determine whether regional occlusion occurs according to the motion monitoring result, and when it is determined that regional occlusion occurs, perform symmetrical completion processing on the adjusted image to obtain a new depth map after symmetrical completion; The new depth map after symmetrical completion is subjected to motion prediction and smooth transition preprocessing, and the focus, foreground and background are separated based on the focus principle, thereby ending the adjustment process.

2. The dual-screen dynamic parallax enhanced immersive naked-eye 3D visual impact display system according to claim 1, characterized in that: The signal control component includes an embedded GPU, a PCB motherboard, and a transmission module. The embedded GPU and the transmission module are configured on the PCB motherboard. The embedded GPU is NVIDIA Jetson, and the transmission module is an optical fiber HDMI or DisplayPort.

3. The dual-screen dynamic parallax enhanced immersive naked-eye 3D visual impact display system according to claim 1, characterized in that: The environmental screen is a Micro-LED screen, the focus screen is a transparent holographic display screen, the acquisition and monitoring module is a depth camera, and the moving module is a stepping motor.

4. The dual-screen dynamic parallax enhanced immersive naked-eye 3D visual impact display system according to claim 1, characterized in that: It also includes a vibration module and an audio module, wherein the input end of the vibration module and the input end of the audio module are electrically connected to the output end of the signal control component, wherein the vibration module adopts an electromagnetic vibration module and an aluminum alloy vibration diaphragm.

5. The dual-screen dynamic parallax enhanced immersive naked-eye 3D visual impact display system according to claim 1, characterized in that: Obtain image data and reference materials to be projected, perform modeling processing on the image data and reference materials, and obtain presentation results, specifically: Obtain image data to be projected, collect reference materials, design based on image data and reference materials, and determine the shape, proportion and dynamic characteristics of image objects; Use Blender to build low-polygon models of the image objects’ shapes, proportions, and dynamic features, define the objects’ overall form and skeleton layout, and optimize their topological structures; Use digital sculpting tools to sculpt low-polygon models in detail, add texture details and undulation details to objects, generate high-resolution normal maps or displacement maps, and unfold the three-dimensional surface into a two-dimensional plane based on UV settings, and use Substance Painter to draw multi-channel texture maps for objects; A skeleton system is added to the low polygon model according to the skeleton binding and weight setting, and the dynamic behavior is recorded based on motion capture, and the obtained data is imported into the rendering engine, and real-time rendering or offline rendering is performed using PBR materials to obtain rendering results, and the rendering results are imported into post-production software to perform color correction, light effect enhancement and sound effect synthesis to obtain presentation results.

6. The dual-screen dynamic parallax enhanced immersive naked-eye 3D visual impact display system according to claim 1, characterized in that: The rendering result is imported into the focus screen and the ambient screen, and the color and lighting between the focus screen and the ambient screen are adjusted, specifically: Importing the 3D model of the presentation result into a focus screen, and importing the background of the presentation result into an environment screen; Color matching and texture mapping algorithms are used to adjust the color and lighting between the focus screen and the ambient screen to ensure that the color changes and lighting directions of the background of the ambient screen and the 3D model of the focus screen are consistent, as well as the color tones and texture styles are consistent.

7. The dual-screen dynamic parallax enhanced immersive naked-eye 3D visual impact display system according to claim 1, characterized in that: When it is determined according to the movement monitoring result that the user's viewing angle has changed, the presentation result is adjusted in the horizontal direction, specifically: Calculate the parallax Δd of the nth frame n , and its calculation formula is: Among them, I p is the distance between the human eye and the 3D model, is the depth of the 3D model in the nth frame, The distance from the user to the screen in the nth frame; According to the parallax amount Δd of the nth frame n Calculate the left eye rendering position of the nth frame and the right eye rendering position of the nth frame The formula is: in, is the original X coordinate of the n-frame of the 3D model; Based on the left eye rendering position and the right eye rendering position, it is assumed that the user's head moves left and right as captured by the monitoring module Move forward and backward When , the parallax amount that needs to be adjusted is: Among them, k x and k z All are dynamic compensation coefficients; Calculate the scaling ratio S of the 3D model after the nth adjustment n , S 0 =S is the original 3D model scaling ratio, α is the perspective deformation compensation coefficient, Δd0 is the parallax of the nth frame, and the presentation result is horizontally adjusted according to the 3D model scaling ratio and the parallax.

8. The dual-screen dynamic parallax-enhanced immersive naked-eye 3D visual impact display system according to claim 1, characterized in that: When it is determined according to the movement monitoring result that the user's viewing angle has changed, the presentation result is adjusted in the vertical direction, specifically: According to the dynamic adjustment formula D f =D0+λ(Z u -Z0) calculates the dynamic adjustment distance between the focus screen and the ambient screen, D0 is the initial distance between the focus screen and the ambient screen in the default state, λ is the adjustment coefficient, Z u is the distance between the user and the screen, and Z0 is the reference user distance; Among them, when it is determined that Z u > Z0, the dynamic adjustment distance D between the focus screen and the ambient screen is increased f , when Z u < Z0, the dynamic adjustment distance D between the focus screen and the ambient screen is decreased f ; Dynamic adjustment distance D between the focus screen and the ambient screen f Based on this, add the user's mobile acceleration and user movement speed Get the updated dynamic adjustment distance D between the focus screen and the ambient screen f =D0+λ(Z u -Z0)+βv u +γa u , where βv u is the dynamic compensation according to the user's moving speed, γa u It is the adjustment value optimized according to the user's acceleration, and β and γ are weight coefficients; Dynamically adjust the distance D between the focus screen and the ambient screen according to the updated f Calculate the displacement of the slide rail s = kD f , the moving module is driven to move the environment screen and the focus screen according to the displacement of the slide rail, wherein k is a slide rail conversion factor.

9. The dual-screen dynamic parallax enhanced immersive naked-eye 3D visual impact display system according to claim 1, characterized in that: According to the motion monitoring result, it is determined whether regional occlusion occurs. When it is determined that regional occlusion occurs, the adjusted image is symmetrically complemented to obtain a new depth map after symmetrical complementation, specifically: Obtain user perspective data (0, φ) according to the mobile monitoring result, and extract pixel depth information Z (x, y) of the 3D model, where θ is the user's current horizontal angle and φ is the user's current pitch angle; The perspective projection matrix P(θ,φ) is calculated based on the user perspective data and pixel depth information to simulate the 3D projection transformation under the user's perspective and determine the area covered by the foreground object. Z forcground (x, y) is the depth of the foreground object, Z background (x, y) is the depth of the background object, where when the depth Z of the foreground object is determined forcground (x,y) is less than the background object depth Z background (x,y), the point is occluded; Calculate the depth gradient between adjacent pixels on the screen And filter out the high frequency area To determine the occlusion boundary, is the partial derivative of Z in the horizontal direction x, indicating the depth change rate of adjacent pixels in the x direction, is the partial derivative of Z in the horizontal direction y, indicating the depth change rate of adjacent pixels in the y direction, and T is the occlusion boundary threshold; The edge of the occluded area is geometrically completed according to the curve interpolation algorithm to obtain the estimated shape of the occluded area, and its mathematical expression is: P(t) = (1-t) 3 P0+3(1-t) 2 tP1+3(1-t)t 2 P2+t 3 P3, and uses Laplace interpolation to expand the edge of the occluded area. Its mathematical expression is: Among them, P0, P1, P2, and P3 are all known boundary points, t is a time parameter whose value is between 0 and 1, ncighbor is the set of four neighboring pixels of the pixel (x, y)(x, y)(x, y), and Z(x i ,y i ) is the neighbor pixel (x i ,y i ), Z(x,y) is the depth value at the screen pixel coordinate (x,y); Calculate the symmetry center of the image C(x,y)=∑Z(x+i,y+j)Z(xi,yj), find the best symmetry axis position coordinates C(x,y), and use the formula Z ′ (x,y)=Z(2C x -x,2C y -y) to complete the symmetric center and obtain a new depth map after symmetric completion, where the coordinates of the symmetric point 2C are used x -x,2C y -y depth value to fill, Z ′ (x,y) is the completed depth value.

10. The dual-screen dynamic parallax enhanced immersive naked-eye 3D visual impact display system according to claim 9, characterized in that: The new depth map after symmetrical completion is subjected to motion prediction and smooth transition preprocessing, and the focus, foreground, and background are separated based on the focus principle, and the adjustment process is ended, specifically: Future perspective of fish conservation based on Kalman prediction algorithm and motion interpolation extreme prediction in, The predicted value of the user's horizontal angle at the next moment. is the predicted value of the user's pitch angle at the next moment, v t is the user's viewing angle velocity, a t is the visual acceleration, ω t is the pitch angle change rate, Δt is the inter-frame time interval; The future position of the key points of the 3D model is calculated according to the Kalman prediction algorithm. The formula is: Among them, X t is the current frame position of the 3D model, V t is the model velocity vector, A t is the model acceleration, R t is the current rotation matrix of the model, Ω t is the model angular velocity, is the predicted value of the position at the next moment, Predict the rotation angle for the next moment; Use motion interpolation algorithm to create smooth transition frames between key frames of 3D model, and use cubic Bézier interpolation to calculate intermediate frames. The formula is: P(t) = (1-t) 3 P0+3(1-t) 2 tP1+3(1-t)t 2 P2+t 3 P3, where P0 and P3 are the positions of the previous and next key frames, and P1 and P2 are control points; According to the formula Seamlessly connect the perspectives of 3D models to ensure smooth movement of the models at different perspectives. left ,X right is the position data of the model at two adjacent perspectives, and u is the interpolation coefficient; Based on the focus principle, an image processing algorithm is used to adjust the dynamic brightness and contrast of the 3D model on the focus screen to separate the focus, foreground and background.

Images