Optical imaging device and image generation method for multiple users

Abstract

The application relates to the field of optical imaging equipment, in particular to optical imaging equipment for multiple users and an image generation method. The optical imaging equipment comprises a server and multiple head-mounted devices, the head-mounted device comprises a display screen, two groups of optical lenses and a fixing mechanism, and the server is provided with an information input unit and an image generation system. With the development of VR technology, more and more fields apply VR technology to stereoscopic display of scenes. In public scenes such as meteorological museums, there are usually many users, and if an image rendering system is equipped for each user, the cost will be greatly increased. The multiple-user optical imaging equipment and the image generation method can realize the image generation requirement of a large number of users by using only one or several image generation systems, and the image can be adjusted for users with different interpupillary distances, so that the image generation quality is ensured and the cost is greatly reduced.

Description

Technical Field

[0001] This invention relates to the field of optical imaging equipment, and more specifically to an optical imaging device and image generation method for multiple users. Background Technology

[0002] Virtual Reality (VR) technology, as a computer simulation system capable of creating and experiencing virtual worlds, has been widely applied in various fields such as entertainment, education, healthcare, industrial design, and cultural tourism in recent years. Its core objective is to create a sense of realism for users through immersive audiovisual experiences. To achieve this goal, VR systems typically rely on head-mounted displays (HMDs), which provide images with parallax to the user's left and right eyes, utilizing the principle of binocular stereoscopic vision to generate three-dimensional spatial perception.

[0003] In current mainstream VR image rendering architectures, a key step in achieving stereoscopic vision is stereoscopic rendering. Specifically, the system needs to independently generate two sets of images with slightly different perspectives for each user—a left-eye view and a right-eye view. These two sets of images are calculated and rendered based on the user's head pose, gaze direction, and interpupillary distance (IPD).

[0004] However, this binocular rendering mechanism based on individualized IPD faces performance and cost challenges in practical applications. Especially for large tourist venues such as exhibition halls, where virtual demonstrations of certain scenes are often required for multiple users simultaneously, using a large surround screen would be extremely costly and severely restrict the venue's functionality. Therefore, simultaneous use of VR by multiple users is a good option. When multiple users simultaneously use VR devices for collaborative experiences (such as virtual museum visits, immersive theaters, or corporate training scenarios), the system must independently execute a complete binocular rendering process for each user. This means that for N users, N sets of image rendering systems are needed to complete N independent image rendering tasks. As the number of users increases, the required number of system sets and the computing power of graphics processing units (GPUs) increase linearly.

[0005] To ensure immersion, VR applications typically require high resolution (e.g., 4K per eye), high refresh rates (90Hz or higher), and low-latency rendering. In multi-user concurrent scenarios, maintaining a high-quality visual experience necessitates high-performance, multi-GPU rendering servers or multiple image rendering devices, which directly leads to a sharp increase in hardware costs, energy consumption, and system complexity. If cost constraints force the use of mid-to-low-end hardware, it becomes necessary to reduce the burden by lowering the resolution, rendering precision, or frame rate.

[0006] Therefore, how to effectively reduce the image rendering load in multi-user VR scenarios while ensuring a personalized visual experience (especially adapting to different IPDs) and overcome the contradiction between performance and cost has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0007] Those skilled in the art should know that distance perception is achieved through images from different perspectives of the two eyes. Since there are significant individual differences in interpupillary distance among different users (usually between 58mm and 72mm), the VR system must dynamically adjust the camera positions of the left and right eyes according to each user's physiological parameters to ensure accurate depth perception of the virtual scene and avoid problems such as visual fatigue, dizziness, or image distortion caused by IPD mismatch.

[0008] The images generated by the image rendering system for people with different interpupillary distances will have a certain angular deviation, which cannot be directly corrected by the matrix mapping of pixels on the display.

[0009] To address this problem, the present invention provides an optical imaging device for multiple users. The optical imaging device includes a server and multiple head-mounted devices. Each head-mounted device includes a display screen, two sets of optical lenses, and a fixing mechanism. The server is equipped with an information input unit and an image generation system.

[0010] The fixing mechanism has a housing, a bracket and a fixing strap. The display screen and optical lens are installed inside the housing. At least two light outlets are provided on the inner side of the housing. Each set of optical lenses is located at one light outlet position. The display screen is located inside the optical lens.

[0011] The bracket is fixedly connected to the outer shell, and the fixing strap is connected to the end of the bracket to fix the outer shell to the user's head;

[0012] The information input unit communicates with the image generation system to receive the measured interpupillary distance of the target user.

[0013] The image generation system generates display images for each user's left and right eyes based on the distribution of interpupillary distances among all users and the interpupillary distances of each user.

[0014] Furthermore, the image generation system includes an image rendering unit and a multi-user image generation unit, wherein the image rendering unit is used to generate rendered images for users with a predetermined interpupillary distance.

[0015] The image rendering unit pre-renders and generates several high-definition simulated scene images based on the existing scene model in the following manner: (1) It receives the interpupillary distance values ​​of all current users, determines the distribution of the interpupillary distance values ​​of all current users, and calculates the interpupillary distance distribution curve of all users, and applies the Gaussian distribution function to the interpupillary distance distribution curve. Perform fitting and determine the fitted μ value. Value, where μ represents the mean interpupillary distance. The standard deviation of the pupillary distance distribution curve is represented by x, which is the abscissa of the Gaussian curve.

[0016] (2) Based on the existing scene model, render and generate: for interpupillary distance of Left and right eye images of the user, left and right eye images of the user with an interpupillary distance of μ, and images of the user with an interpupillary distance of μ. Images of the user's left and right eyes;

[0017] The multi-user image generation unit generates a left-eye image for each user based on their actual interpupillary distance, using at least two of the three rendered left-eye images, and generates a right-eye image for the user based on at least two of the three rendered right-eye images. This process includes:

[0018] (3) For each user, calculate the horizontal distance W2 from the center of the screen to each pixel on the display screen. W 总 x0 is the total width of the screen, x2 is the total number of pixels horizontally, and x3 is the horizontal coordinate of the pixel on the screen.

[0019] (4) Determine the user's interpupillary distance μ2, and calculate the lateral deflection angle of each pixel relative to the center line directly in front of the eyeball at the current interpupillary distance. Where D1 is the distance from the eyeball to the display screen, based on this deflection angle, it is substituted into the user's perspective with an interpupillary distance of μ for inversion to determine the pixel position observed by the user with an interpupillary distance of μ at this deflection angle and the corresponding pixel value in the image; substituted into the interpupillary distance of μ... The inversion is performed from the user's perspective to determine the interpupillary distance at that deflection angle. The pixel positions observed by the user and the corresponding pixel values ​​in the image are substituted into the interpupillary distance. The inversion is performed from the user's perspective to determine the interpupillary distance at that deflection angle. The pixel positions observed by the user and the corresponding pixel values ​​in the image, based on each pixel's position in the image. μ and At least two of the pixel values ​​of the image corresponding to the interpupillary distance are weighted to obtain the pixel value of each pixel in the image at the current interpupillary distance.

[0020] Furthermore, the optical imaging device also includes a sound playback device for playing sound.

[0021] On the other hand, the present invention provides an image generation method for multiple users, the method comprising: (1) measuring the interpupillary distance values ​​of all current users respectively, determining the distribution of all interpupillary distance values ​​of the current users, and plotting the interpupillary distance distribution curves of all users, and applying a Gaussian distribution function to the user distribution curves. Perform fitting and determine the fitted μ value. Value, where μ represents the mean interpupillary distance. The standard deviation of the distribution curve is represented by x, where x is the x-axis of the Gaussian curve.

[0022] (2) Based on the existing scene model, render and generate the scene for interpupillary distance of . The user's left and right eye images, the user's left and right eye images for a pupillary distance of μ, and the user's pupillary distance for a distance of μ. Images of the user's left and right eyes;

[0023] The multi-user image generation unit generates left and right eye images for each user based on their actual interpupillary distance:

[0024] The process includes:

[0025] (3) For each user, determine the user's interpupillary distance μ2, and calculate the horizontal distance W2 of each pixel on the display screen from the center of the screen at the current interpupillary distance. W 总 x0 is the total width of the screen, x2 is the total number of pixels horizontally, and x3 is the horizontal coordinate of the pixel on the screen.

[0026] (4) Calculate the lateral deflection angle of each pixel relative to the center line directly in front of the eyeball at the current interpupillary distance. Based on this deflection angle, the image is inverted using the user's perspective with an interpupillary distance of μ to determine the screen pixel position and pixel value observed by the user with an interpupillary distance of μ at this deflection angle; then, the image is inverted using the user's perspective with an interpupillary distance of μ. The inversion is performed from the user's perspective to determine the interpupillary distance at that deflection angle. The pixel positions observed by the user and the corresponding pixel values ​​in the image are substituted into the interpupillary distance. The inversion is performed from the user's perspective to determine the interpupillary distance at that deflection angle. The screen pixel positions and pixel values ​​observed by the user, based on each pixel's position. μ and The pixel values ​​of each pixel in the current interpupillary distance image are obtained by weighting at least two of the pixel values ​​of the corresponding pixel values ​​in the image.

[0027] Further, the method includes: retrieving several target user images initially generated by the multi-user image generation unit; rendering target user images with the same interpupillary distance using an image rendering unit; verifying the image generated by the multi-user image generation unit using the rendered image; calculating the sum of the pixel value variances of all pixels in both images; if the sum of the pixel value variances of all pixels is less than a preset threshold, the generated image is deemed qualified; otherwise, the method adjusts the image generation process of the user image generation unit. μ and The weighted weights of the image corresponding to the interpupillary distance.

[0028] Beneficial effects

[0029] This invention enables the generation of high-definition rendered images for users with different interpupillary distances with only one or a few image rendering systems, even when used by a large number of users simultaneously. This solves the batch demand for high-definition imaging and display in science popularization venues such as meteorological museums, and can also significantly reduce costs. Attached Figure Description

[0030] Figure 1 This is a schematic flowchart of the image generation method of the present invention;

[0031] Figure 2 This is a schematic diagram of the structure of the head-mounted device in the optical imaging apparatus of the present invention;

[0032] Figure 3 This is a schematic diagram showing the positional relationship between the display screen and the optical lens in the optical imaging device of the present invention. The display screen can be a single display screen divided into half for display, or two display screens can be used.

[0033] Figure 4 This is a view showing the relationship between the viewing angle of a user with an interpupillary distance of μ and the screen display, using the left eye as an example.

[0034] Figure 5 This is a view showing the relationship between the viewing angle of a user with an interpupillary distance of μ2 and the screen display, using the left eye as an example.

[0035] Figure 6 To be Figure 5 A schematic diagram showing the α2 angle inversion back to the observation angle of a user with an interpupillary distance of μ. Detailed Implementation

[0036] The specific implementation of the present invention will be described in further detail below with reference to the accompanying drawings. Those skilled in the art should understand that the following embodiments are merely examples and are not intended to limit the scope of protection of the present invention.

[0037] First, let me introduce the application scenarios of this invention. This invention is mainly used for popular science education or other scenarios where multiple users can simultaneously access and display information to the public. For example, it can be applied in popular science settings such as meteorological museums and planetariums.

[0038] In use, a server and multiple head-mounted devices are first provided. Each head-mounted device is assigned to a user. Before all users put on the head-mounted devices of this invention, the staff first measures their interpupillary distance and inputs the interpupillary distance value and device number into the server through the information entry unit.

[0039] The method of the present invention is as follows Figure 1 As shown, in summary, the method of the present invention includes the following steps: plotting the interpupillary distance distribution curves for all users and fitting them using a Gaussian function; determining the interpupillary distance of the target user for each user; calculating the deflection angle of each pixel relative to the center line directly in front of the eyeball at the current interpupillary distance; and using the obtained deflection angle of each pixel to perform inversion and weighted summation to obtain the pixel value of each pixel in the current interpupillary distance image. The method of this embodiment will be further described in detail below with examples of specific devices.

[0040] like Figure 2-3 As shown, the head-mounted device of this embodiment includes: a display screen 1 (one or two screens), two sets of optical lenses 2, and a fixing mechanism. The server is not shown in the figure; the server includes an information input unit and an image generation system.

[0041] The fixing mechanism has a housing 31, a bracket 32, and a fixing strap 33. The bracket 32 ​​is used to facilitate connection and fixation to the housing 31. The fixing strap 33 is connected to the end of the bracket and is used to fix the entire structure to the user's head. Preferably, the fixing strap 33 is adjustable in length or is an elastic fixing strap.

[0042] The display screen 1 and the optical lens 2 are installed inside the housing, and at least two light outlets are provided inside the housing, such as... Figure 3 As shown, each group of optical lenses is located at a light outlet position, and the display screen is located inside the optical lenses. Here, a non-curvature head-up lens is used, which is only used for the light outlet. Those skilled in the art can customize a curvature lens with diopter according to the user's needs.

[0043] The server receives the interpupillary distance (IPD) values ​​of all current users and, based on these values, determines the distribution of all IPD values ​​for each user. It then identifies the median, maximum, and minimum IPD values ​​and calculates the IPD distribution curve for all users, applying a Gaussian distribution function to the curves. The system performs a fitting process to determine the fitted μ and σ values, where μ represents the mean interpupillary distance (IPD) and σ represents the standard deviation of the distribution curve. This method allows us to determine the IPD distribution of the current batch of users. For example, if the current batch consists of elementary school students, their IPDs will generally be smaller; if it consists of adults, their IPDs will generally be larger.

[0044] The server then renders and generates the images. Image rendering is achieved using Unreal Engine VR or Unity XR. Of course, those skilled in the art can also use other virtual reality simulation software to generate the rendered images. In the image rendering software, only the interpupillary distance parameter needs to be adjusted to generate images with different interpupillary distances.

[0045] For example, using Unreal Engine VR software, there are numerous weather and environment plugins available in the Unreal Engine marketplace. Load these plugins into the software, enable VR support mode, create a project, and select a VR template. Retrieve a pre-recorded or rendered stereoscopic scene video model from Unreal Engine VR, determine the primary viewing position for that scene, and the binocular camera spacing (i.e., left and right eye spacing) for that position. The left eye's viewpoint (left camera) is located at a predetermined distance (e.g., μ / 2) to the left of the primary viewing position, and a display image for the left eye is generated. Similarly, the right eye's viewpoint (right camera) is located at a predetermined distance (e.g., μ / 2) to the right of the primary viewing position, and a display image for the right eye is generated. Then, export the images. More preferably, scene variation parameters can be further added to the stereoscopic scene video model, such as using Enviro Sky and Weather to adjust scene effects like rainfall intensity.

[0046] Based on the existing scene model, generate at least three left-eye images and three right-eye images, with the three left-eye images P... L1 P L2 P L3 The corresponding interpupillary distances for the users are respectively ,μ, Left-eye rendered images in three different scenarios; three right-eye images P R1 P R2 P R3 The corresponding interpupillary distances for the users are respectively ,μ, The right eye rendered image in three cases (the distance between the center of the generated binocular images on the screen is as described above).

[0047] In another preferred implementation, five images can be rendered, each corresponding to the user's interpupillary distance. , ,μ, , Left and right eye images in five different scenarios.

[0048] According to the Gaussian distribution law This can cover 68% of users, while It can cover 95% of users.

[0049] This method allows as many users as possible to be covered within the rendered image.

[0050] The following example, using the left eye image, illustrates how to present images obtained from a single rendering to different users.

[0051] For the three left-eye images P L1 P L2 P L3 , where P L1 For interpupillary distance of User-rendered generation, P L2 P is generated for rendering for users with an interpupillary distance of μ. L3 For interpupillary distance of User-rendered.

[0052] Figure 4 This is a diagram illustrating the simplest standard case, specifically the lateral viewing direction from the center of the eye O when viewing any point Q1 on the screen. For example... Figure 4 As shown, for a pupillary distance of μ, the position directly opposite the center of the eyeball O (calculated here as the center of the lens) is set as the center of the screen O1 (corresponding to the viewpoint center of the rendered image), and its distance from the screen is D1 (this distance can be approximated beforehand by measuring the average distance of various users wearing the device). The distance from point Q1 on the screen to the center of the screen is Q1O1, denoted as W1. Because there are the most users near the pupillary distance μ, the image presentation is set with the position directly opposite the user with a pupillary distance of μ as the center of the screen O1. The entire image seen from this perspective is the left eye image P. L2 .

[0053] For any one user among many users, but whose interpupillary distance is not always μ, let's assume their interpupillary distance is μ2, in Within the specified range, the output process for the user's left eye image is as follows:

[0054] (1) Determine the estimated distance D1 between the center of the wearer's eyeball (based on the lens) and the center of the left eye display screen, i.e. Figure 4-5 The distance between O1 and O2 is shown in both figures. Both figures are from a top-down view and show the relationship between the observation direction and the screen. The right side (the plane where O1 and O2 are located) is the screen.

[0055] (2) For the left eye image P L1 P L2 P L3 For each image in the image, using a plane that is vertical and passes through the center line of the eye's line of sight as the reference plane, calculate the angle between the ray formed by the center point of the eyeball as the first point O and any point on the screen as the second point and the reference plane, to determine the preset horizontal viewing angle of the human eye for each pixel under the preset interpupillary distance of the current image.

[0056] (3) Determine the interpupillary distance of the target user and calculate Q2(x2,y2) of each pixel on the display screen for the target user's current interpupillary distance.

[0057] like Figure 5 As shown, for a certain user, their corresponding interpupillary distance is μ2 < μ (assuming the user's interpupillary distance is less than μ). Since the current user's interpupillary distance is less than μ, the distance (O2Q2) between the center of their eyeball O' (calculated here using the center of the lens) and the midpoint between their eyes is μ2 / 2. This causes the position of the center of the eyeball O to shift to the right (lower part of the figure) relative to a user with an interpupillary distance of μ. The shift distance is the distance between O1 and O2, which is... If the user's interpupillary distance is greater than μ, then A negative value indicates an offset in the opposite direction. Assuming pixel coordinates are Q2(x2, y2), and its horizontal distance from the screen center O1 on the left-eye side of the display screen is W2, This is the lateral distance of the pixel from the center line O2 of the eyeball. W 总 The total width of the screen is x0, which is the total number of pixels horizontally, i.e., the total number of columns. The maximum pixel value is set at the bottom. If W2 is positive, it means that the pixel is located in the right half of the screen (bottom of the figure). If it is negative, it is located in the left half of the screen (top of the figure).

[0058] The angle between O'Q2 and O'O2 is Since the change in interpupillary distance does not affect the vertical observation direction, the horizontal viewing angle of the human eye corresponding to each column of pixels in the horizontal direction is the same. Therefore, the expected horizontal viewing angle of each pixel in the current image in the eyes of a person with an interpupillary distance of μ can be determined.

[0059] Based on the aforementioned angle, i.e., the lateral deflection angle of a pixel relative to the central plane of the eyeball, it is substituted into the user's visual field with an interpupillary distance of μ for inversion, determining the screen pixel position and pixel value observed by the user with an interpupillary distance of μ at this deflection angle. The aforementioned angle is then substituted into... Figure 4 In the positional relationship shown, the result is as follows Figure 6 As shown, the pixel coordinates corresponding to D1×tanα2 can be determined, and the pixel value can be determined based on these pixel coordinates Q3; similarly, based on this deflection angle, the interpupillary distance is substituted... The inversion is performed from the user's perspective to determine the interpupillary distance at that deflection angle. Based on α2, the pixel position and pixel value observed by the user on the screen can be used to determine the pixel point corresponding to that angle in the image rendered with that viewpoint as the center, with its horizontal coordinate being... This allows us to determine the pixel value; based on the pixel's position in μ and... The pixel values ​​of each pixel in the current interpupillary distance image are obtained by weighting the pixel values ​​of the corresponding pixels in the image.

[0060] Preferably, the horizontal and vertical gradients of the pixel values ​​of each pixel at the determined current interpupillary distance are calculated to determine the gradient rate of change of pixel values ​​between each pixel at the current interpupillary distance. This gradient rate of change of pixel values ​​between each pixel at the current interpupillary distance is then compared with the gradients at μ and μ. The gradient change rate of the corresponding position in the image corresponding to the pupil distance is compared to determine whether the gradient change rate of the pixel value between each pixel point under the current pupil distance exceeds the predetermined ratio of the gradient change rates of the two. If it exceeds the ratio, the pixels in the target area are smoothed.

[0061] Using a similar approach, for each user, determine which of the following categories their interpupillary distance falls into: (1) less than (2) (3) In Interval, (4) greater than For cases (1) and (2), the rendered image with an interpupillary distance of μ and the interpupillary distance of μ are used. The rendered image is weighted and inverted to obtain the rendered image of the target user. For cases (3) and (4), a rendered image with an interpupillary distance of μ and an interpupillary distance of μ are used. The rendered image is weighted and inverted to obtain the rendered image for the target user. Taking the latter case as an example... k1, k2, and k3 are weighting coefficients, L2 is the weighted pixel value, and L... μ The pixel values ​​of the rendered image with interpupillary distance μ at the same viewpoint and The interpupillary distance is The pixel values ​​in the rendered image at the same viewpoint.

[0062] This invention is described using a non-curvature lens as an example. If the optical lens has a certain magnification or reduction ratio, then the magnification is adjusted accordingly. The value is transformed in reverse.

[0063] Since the effects of visual observation can only be directly observed by humans to obtain the most direct feedback, to verify the effectiveness of the method of this invention, 50 visitors were invited to experience it for free, and their feedback was collected. The experimental procedure is as follows: After measuring the interpupillary distance of the visitors, the 50 visitors were asked to wear VR devices and watch two meteorological simulation videos. The first video was entirely rendered and generated by an image generation system, while the second video was generated using the method of this invention. The visitors were then asked to fill out a questionnaire, and the questionnaire results were statistically analyzed. The statistical results are shown in the table below:

[0064] Table 1

[0065]

[0066] User surveys show that the vast majority of users are satisfied with the video quality and cannot distinguish between images generated using the method of this invention and images generated by direct rendering. This proves that the images generated by the method of this invention are roughly equivalent in quality to images generated directly for each user and can be widely applied.

[0067] This approach eliminates the need for a separate image rendering and generation system for each user when a large number of users are using the system simultaneously. Instead, it allows for the generation of high-definition rendered images for different interpupillary distances using only one or a few image rendering systems. This addresses the bulk demand for high-definition weather simulation displays in science education venues such as meteorological museums, while also significantly reducing costs.

Claims

1. An optical imaging device for multiple users, characterized in that, The optical imaging equipment includes a server and multiple head-mounted devices. Each head-mounted device includes a display screen, two sets of optical lenses, and a fixing mechanism. The server is equipped with an information input unit and an image generation system. The fixing mechanism has a housing, a bracket and a fixing strap. The display screen and optical lens are installed inside the housing. At least two light outlets are provided on the inner side of the housing. Each set of optical lenses is located at one light outlet position. The display screen is located inside the optical lens. The bracket is fixedly connected to the outer shell, and the fixing strap is provided at the end of the bracket to fix the outer shell to the user's head; The information input unit communicates with the image generation system to receive the measured interpupillary distance of the target user. The image generation system generates display images for each user's left and right eyes based on the distribution of interpupillary distances among all users and the interpupillary distances of each user. The image generation system includes an image rendering unit and a multi-user image generation unit. The image rendering unit is used to generate rendered images for users with a predetermined interpupillary distance. The image rendering unit pre-renders and generates several high-definition simulated scene images based on the existing scene model in the following manner: (1) It receives the interpupillary distance values ​​of all current users, determines the distribution of the interpupillary distance values ​​of all current users, and calculates the interpupillary distance distribution curve of all users, and applies the Gaussian distribution function to the interpupillary distance distribution curve. Perform fitting and determine the fitted μ value. Value, where μ represents the mean interpupillary distance. The standard deviation of the pupillary distance distribution curve is represented by x, which is the abscissa of the Gaussian curve. (2) Based on the existing scene model, render and generate: for interpupillary distance of Left and right eye images of the user, left and right eye images of the user with an interpupillary distance of μ, and images of the user with an interpupillary distance of μ. Images of the user's left and right eyes; The multi-user image generation unit generates a left-eye image for each user based on their actual interpupillary distance, using at least two of the three rendered left-eye images, and generates a right-eye image for the user based on at least two of the three rendered right-eye images. This process includes: (3) For each image of each user, calculate the horizontal distance W2 of each pixel on the display screen from the center of the screen. W 总 x0 is the total width of the screen, x2 is the total number of pixels horizontally, and x3 is the horizontal coordinate of the pixel on the screen. (4) Determine the user's interpupillary distance μ2, and calculate the lateral deflection angle of each pixel relative to the center line directly in front of the eyeball at the current interpupillary distance. Where D1 is the distance from the eyeball to the display screen, based on this deflection angle, it is substituted into the user's perspective with an interpupillary distance of μ for inversion to determine the pixel position observed by the user with an interpupillary distance of μ at this deflection angle and the corresponding pixel value in the image; substituted into the interpupillary distance of μ... The inversion is performed from the user's perspective to determine the interpupillary distance at that deflection angle. The pixel positions observed by the user and the corresponding pixel values ​​in the image are substituted into the interpupillary distance. The inversion is performed from the user's perspective to determine the interpupillary distance at that deflection angle. The pixel positions observed by the user and the corresponding pixel values ​​in the image, based on each pixel's position in the image. μ and At least two of the pixel values ​​of the corresponding pixel points in the image are weighted to obtain the pixel values ​​of each pixel point in the image at the current pupil distance.

2. The optical imaging device for multiple users according to claim 1, characterized in that, It also includes a sound playback device for playing sound.

3. A method for generating images for multiple users, characterized in that, The method includes: (1) measuring the pupillary distance values ​​of all current users respectively, determining the distribution of all pupillary distance values ​​of the current users, and plotting the pupillary distance distribution curves of all users, and applying the Gaussian distribution function to the user distribution curves. Perform fitting and determine the fitted μ value. Value, where μ represents the mean interpupillary distance. The standard deviation of the distribution curve is represented by x, where x is the x-axis of the Gaussian curve. (2) Based on the existing scene model, render and generate the scene for interpupillary distance of . The user's left and right eye images, the user's left and right eye images for a pupillary distance of μ, and the user's pupillary distance for a distance of μ. Images of the user's left and right eyes; Using a multi-user image generation unit, a left-eye image is generated for each user based on their actual interpupillary distance, using at least two of the three rendered left-eye images, and a right-eye image is generated for the user based on at least two of the three rendered right-eye images. The process includes: (3) For each user, determine the user's interpupillary distance μ2, and calculate the horizontal distance W2 of each pixel on the display screen from the center of the screen at the current interpupillary distance. W 总 x0 is the total width of the screen, x2 is the total number of pixels horizontally, and x3 is the horizontal coordinate of the pixel on the screen. (4) Calculate the lateral deflection angle of each pixel relative to the center line directly in front of the eyeball at the current interpupillary distance. Where D1 is the distance from the eyeball to the display screen, based on this deflection angle, it is substituted into the user's perspective with an interpupillary distance of μ for inversion to determine the screen pixel position and pixel value observed by the user with an interpupillary distance of μ at this deflection angle; substituted into the interpupillary distance of μ... The inversion is performed from the user's perspective to determine the interpupillary distance at that deflection angle. The pixel positions observed by the user and the corresponding pixel values ​​in the image are substituted into the interpupillary distance. The inversion is performed from the user's perspective to determine the interpupillary distance at that deflection angle. The screen pixel positions and pixel values ​​observed by the user, based on each pixel's position. μ and At least two of the pixel values ​​of the corresponding pixel points in the image are weighted to obtain the pixel values ​​of each pixel point in the current interpupillary distance image.

4. The image generation method for multiple users according to claim 3, characterized in that, The method includes: retrieving several user images initially generated by a multi-user image generation unit; rendering user images with the same interpupillary distance using an image rendering unit; verifying the image generated by the multi-user image generation unit using the rendered image; calculating the sum of the pixel value variances of all pixels in both images; if the sum of the pixel value variances of all pixels is less than a preset threshold, the generated image is deemed qualified; otherwise, the method adjusts the image generation process of the user image generation unit. μ and The weighted weights of the image corresponding to the interpupillary distance.

Images