Zero-parallax plane adjustment method, device, equipment and storage medium

Abstract

The application provides a zero-parallax plane adjusting method, device, equipment and storage medium. The method comprises the following steps: determining a reference zero-parallax plane of a three-dimensional scene model; determining a focusing area corresponding to the reference zero-parallax plane and highlighting the focusing area corresponding to the reference zero-parallax plane; in response to a user's adjustment operation on the position of the zero-parallax plane, determining a target zero-parallax plane according to the adjustment operation and the reference zero-parallax plane, and determining a focusing area corresponding to the target zero-parallax plane; and updating the highlighted focusing area to the focusing area corresponding to the target zero-parallax plane. The application can improve the interactivity and viewing experience of naked-eye 3D display.

Description

Technical Field

[0001] This application relates to the field of naked-eye 3D display technology, and in particular to a zero parallax plane adjustment method, apparatus, device, and storage medium. Background Technology

[0002] Glasses-free 3D display technology can present viewers with a sense of depth and stereoscopic images without the need for auxiliary devices. In stereoscopic displays, the scene content can form in-screen and out-of-screen areas relative to the display plane. The boundary between in-screen and out-of-screen areas is usually called the zero parallax plane (also known as the "zero plane").

[0003] In glasses-free 3D displays, the position of the zero-parallax plane affects the degree to which the stereoscopic image extends beyond the screen and into the screen. The zero-parallax plane in related technologies is usually determined automatically by algorithms. However, due to significant differences in depth distribution across different scenes and varying user preferences for stereoscopic effects, the zero-parallax plane automatically determined by the algorithm often does not match user expectations, thus impacting the viewing experience of glasses-free 3D displays. Summary of the Invention

[0004] In view of this, embodiments of this application provide a zero parallax plane adjustment method, apparatus, device, and storage medium to solve at least one problem existing in the prior art.

[0005] In a first aspect, embodiments of this application provide a zero-parallax plane adjustment method, the method comprising: Determine the reference zero-parallax plane for the 3D scene model; Determine the focus area corresponding to the reference zero parallax plane and highlight the focus area corresponding to the reference zero parallax plane; In response to the user's adjustment operation on the position of the zero parallax plane, the target zero parallax plane is determined based on the adjustment operation and the reference zero parallax plane, and the focus area corresponding to the target zero parallax plane is determined. The highlighted focus area is updated to the focus area corresponding to the target zero parallax plane.

[0006] Secondly, embodiments of this application provide a zero parallax plane adjustment device, the device comprising: The first determining module is used to determine the reference zero parallax plane of the three-dimensional scene model and to determine the focus area corresponding to the reference zero parallax plane. The high-brightness display module is used to highlight the focus area corresponding to the reference zero-parallax plane; The zero parallax adjustment module is used to respond to the user's adjustment operation on the position of the zero parallax plane, and determine the target zero parallax plane based on the adjustment operation and the reference zero parallax plane. The second determining module is used to determine the focus area corresponding to the target zero parallax plane; The display module is updated to update the highlighted focus area to the focus area corresponding to the target zero parallax plane.

[0007] Thirdly, embodiments of this application provide an electronic device including a processor, a memory, and an executable program stored in the memory and executable by the processor, wherein the processor executes the steps of the zero parallax plane adjustment method provided in the first aspect when running the executable program.

[0008] Fourthly, embodiments of this application provide a storage medium storing an executable program thereon, which, when executed by a processor, implements the steps of the zero parallax plane adjustment method provided in the first aspect.

[0009] This application provides a method, apparatus, device, and storage medium for adjusting a zero-parallax plane. By determining a reference zero-parallax plane and its corresponding focus area in a 3D scene model, and highlighting the determined focus area, the user can intuitively see the position of the reference zero-parallax plane in the 3D scene. In response to the user's adjustment operation on the position of the zero-parallax plane, a target zero-parallax plane is determined based on the adjustment operation and the reference zero-parallax plane, and the focus area corresponding to the target zero-parallax plane is determined. The highlighted focus area is then updated to match the focus area corresponding to the target zero-parallax plane. This allows the user to adjust the zero-parallax plane according to their viewing preferences. The zero-parallax plane and its highlighted focus area in the 3D scene change synchronously with the adjustment operation, thereby achieving an out-of-screen or in-screen effect that better suits the user's individual needs, thus improving the interactivity and viewing experience of naked-eye 3D display. Attached Figure Description

[0010] Figure 1 This is one of the flowcharts illustrating the zero parallax plane adjustment method provided in the embodiments of this application; Figure 2 A second schematic flowchart illustrating the zero parallax plane adjustment method provided in this application embodiment; Figure 3 The third schematic flowchart of the zero parallax plane adjustment method provided in the embodiments of this application; Figure 4 Fourth flowchart illustrating the zero parallax plane adjustment method provided in this application embodiment; Figure 5 Fifth flowchart illustrating the zero parallax plane adjustment method provided in the embodiments of this application; Figure 6a A schematic diagram of the interface when the zero parallax plane is located in the center of the scene, as provided in an embodiment of this application. Figure 6b A schematic diagram of the interface when the zero parallax plane is located near the scene, as provided in the embodiments of this application; Figure 6c A schematic diagram of the interface when the zero parallax plane is located at a distance in the scene, as provided in an embodiment of this application; Figure 6d This is one of the schematic diagrams of highlight color distribution in a three-dimensional scene provided in the embodiments of this application; Figure 6e This is the second schematic diagram of the distribution of highlight colors in a three-dimensional scene provided in the embodiments of this application; Figure 6f This is the third schematic diagram of the distribution of highlight colors in a three-dimensional scene provided in the embodiments of this application; Figure 7 This is a schematic diagram of the zero parallax plane adjustment device provided in an embodiment of this application. Detailed Implementation

[0011] To make the technical solution and beneficial effects of this application more apparent and understandable, a detailed description is provided below by listing specific embodiments. The accompanying drawings are not necessarily drawn to scale, and local features may be enlarged or reduced to more clearly show the details of the local features; unless otherwise defined, the technical and scientific terms used herein have the same meanings as those in the technical field to which this application pertains.

[0012] The embodiments in this application are not exhaustive, but merely illustrative of some embodiments, and are not intended to limit the scope of protection of this disclosure. Unless otherwise specified, each step in a particular embodiment can be implemented as an independent embodiment, and the steps can be arbitrarily combined. For example, a solution after removing some steps in a particular embodiment can also be implemented as an independent embodiment, and the order of the steps in a particular embodiment can be arbitrarily interchanged. Furthermore, the optional implementation methods in a particular embodiment can be arbitrarily combined; moreover, the embodiments can be arbitrarily combined, for example, some or all steps of different embodiments can be arbitrarily combined, and a particular embodiment can be arbitrarily combined with the optional implementation methods of other embodiments.

[0013] In each embodiment of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of the embodiments are consistent and can be referenced by each other. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0014] In the description of the embodiments of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0015] To facilitate understanding, a brief explanation of some of the terms used in this application will be provided first.

[0016] Naked-eye 3D display: refers to a display technology that does not require external auxiliary devices such as glasses. It achieves a multi-viewpoint stereoscopic display effect by setting optical structures such as lenticular lenses, gratings or microlens arrays in front of the display panel, so that different pixels or sub-pixels on the display panel emit light in different spatial directions, thereby forming multiple observable viewpoints in space.

[0017] Spatial primitives: 3D Gaussian ellipsoids, point cloud points, or other scene units with a 3D center position that can participate in multi-view rendering. For example, using a Gaussian ellipsoid as a spatial primitive, its 3D center position can be the center point of the Gaussian ellipsoid in a 3DGS (3D Gaussian Splatting) model PLY (Polygon File Format) file.

[0018] Camera spatial depth: The projection of the spatial primitive world coordinates onto the forward unit vector of the camera relative to the observation camera position.

[0019] Zero parallax plane: A spatial plane at which all off-axis viewpoints converge and where lateral parallax is zero, which can be used as the boundary between the out-of-screen zone and the in-screen zone.

[0020] Hardware optical viewpoints: Multiple actual observation directions formed by the terminal's directional optical structure; different from the base camera viewpoint (i.e., the base rendering viewpoint) of the actual software rendering.

[0021] Peak focus display: Calculate the superposition intensity of a specific color based on the absolute depth difference from the spatial primitive to the zero parallax plane, so that a bright peak band is formed near the zero parallax plane.

[0022] Round trip phase sequence: A periodic viewpoint index sequence formed by N basic rendering viewpoints. This sequence starts from 0, increases sequentially to N-1, and then decreases sequentially back to 0. That is, the length of the round trip phase sequence is M=2N-1.

[0023] Naked-eye 3D display terminals use lenticular lenses, gratings, or other directional optical structures to direct different pixels or color sub-pixels of the display panel towards different viewing directions. After a 2D image is transformed into a 3D scene representation, a zero-parallax plane needs to be determined, multiple viewing points need to be generated, and these points need to be interwoven according to the terminal's optical parameters. The position of the zero-parallax plane affects the degree to which the stereoscopic image appears off-screen and in-screen. In related technologies, the zero-parallax plane is usually determined automatically by algorithms. However, due to significant differences in depth distribution across different scenes and varying user preferences for stereoscopic effects, the automatically determined zero-parallax plane often does not match user expectations. For example, some users want the entire scene to be behind the screen (i.e., completely in-screen), creating an effect similar to viewing through a window; others want all content to protrude from the screen, creating a strong off-screen effect; still others want part of the scene to appear off-screen and part to appear in-screen to enhance the sense of depth. When the position of the zero-parallax plane does not meet user expectations, users lack intuitive and convenient means to adjust it. This negatively impacts the viewing experience of naked-eye 3D displays.

[0024] Figure 1 This is a schematic flowchart illustrating the zero parallax plane adjustment method provided in an embodiment of this application. Figure 1 As shown, the method includes the following steps: S101: Determine the reference zero-parallax plane for the 3D scene model; S102: Determine the focus area corresponding to the reference zero parallax plane and highlight the focus area corresponding to the reference zero parallax plane; S103: In response to the user's adjustment operation on the position of the zero parallax plane, determine the target zero parallax plane based on the adjustment operation and the reference zero parallax plane, and determine the focus area corresponding to the target zero parallax plane; S104: Update the highlighted focus area to the focus area corresponding to the target zero parallax plane.

[0025] In this embodiment, the zero-parallax plane adjustment method can be executed by a zero-parallax plane adjustment device. The zero-parallax plane adjustment device can be implemented in hardware, software, or a combination thereof, and can be configured in a glasses-free 3D display device or a terminal with glasses-free 3D display functionality. The terminal may include, but is not limited to, mobile phones, tablets, personal computers, advertising screens, and exhibition display equipment. Users can input adjustment operations via touchscreen, mouse, remote control, or gesture recognition to adjust the zero-parallax plane.

[0026] In step S101, the three-dimensional scene model can be obtained based on a two-dimensional image. For example, the three-dimensional scene model can be generated based on a two-dimensional image using a three-dimensional scene generation program or a three-dimensional reconstruction algorithm. The two-dimensional image can be a monocular image, a binocular image, a multi-view image, or a video frame, etc.

[0027] The 3D scene model comprises multiple spatial primitives. These spatial primitives can be 3D Gaussian primitives (i.e., Gaussian ellipsoids). In 3DGS representation, the 3D scene model can store the attribute parameters of all 3D Gaussian primitives in the scene through a PLY file; where each vertex represents a 3D Gaussian primitive, and its attribute parameters can include the center position, covariance, opacity, and spherical harmonic coefficients of that 3D Gaussian primitive.

[0028] In some examples, a reference zero-parallax plane can be determined based on the camera space depth of each spatial primitive in a 3D scene model. The process of obtaining the camera space depth of the spatial primitives may include: determining the center position of each spatial primitive, and transforming each center position to the world coordinate system according to the entity world transformation matrix to obtain the camera space depth of each spatial primitive. By calculating the depth statistics of each center position in camera space, such as average depth, median depth, or weighted depth, the plane corresponding to the depth statistics is determined as the reference zero-parallax plane.

[0029] In other examples, the reference zero-parallax plane can be directly determined using the center of the scene bounding box. For instance, the depth of the center of the bounding box of all spatial primitives in the 3D scene model in camera space can be calculated as the depth of the reference zero-parallax plane. The semi-axis length of the bounding box in the depth direction can be used to estimate the distribution range of the scene in the depth direction.

[0030] In step S102, the spatial range that meets the preset conditions in the depth direction can be determined based on the reference zero parallax plane as the focus area corresponding to the reference zero parallax plane; wherein, the preset conditions include, for example, the absolute depth difference between the center position of the spatial primitive and the reference zero parallax plane not exceeding a preset depth threshold.

[0031] The focus area can be highlighted, for example, by rendering all spatial primitives within the focus area in green or another striking color. Here, any spatial primitive within the focus area means that the center of that spatial primitive falls within the focus area. Thus, by highlighting the focus area corresponding to the reference zero-parallax plane, the focus area corresponding to the reference zero-parallax plane can be displayed intuitively, allowing users to directly perceive the position of the reference zero-parallax plane in the 3D scene.

[0032] When the user adjusts the position of the zero parallax plane on the display interface, the zero parallax plane adjustment device changes the position of the zero parallax plane in the depth direction according to the adjustment operation to obtain the target zero parallax plane, and simultaneously determines the focus area corresponding to the target zero parallax plane, and updates the highlighted focus area to the focus area corresponding to the target zero parallax plane.

[0033] It is understood that users can input the adjustment operation through preset sliders, levers or other controls on the interactive interface.

[0034] In step S103 above, the focus area corresponding to the zero-parallax plane of the target can be determined based on the depth difference between each spatial primitive in the three-dimensional scene model and the target zero-parallax plane.

[0035] The specific implementation method for determining the focus area corresponding to the target zero parallax plane can refer to the implementation method for determining the focus area corresponding to the reference zero parallax plane in the aforementioned step S102. It is only necessary to replace the reference zero parallax plane with the target zero parallax plane, which will not be elaborated here.

[0036] In step S104 above, the highlight display of the focus area corresponding to the zero parallax plane before adjustment is canceled, and the focus area corresponding to the target zero parallax plane is highlighted.

[0037] The method of highlighting the focus area corresponding to the target zero parallax plane can be the same as the method of highlighting the focus area corresponding to the reference zero parallax plane in the aforementioned step S102. It is only necessary to replace the reference zero parallax plane with the target zero parallax plane, which will not be described in detail here.

[0038] In this embodiment, users can adjust the zero parallax plane to their desired position according to their viewing preferences. During the adjustment process, the highlighted areas in the 3D scene change synchronously with the adjustment operation, allowing users to intuitively see the highlighted areas change as the zero parallax plane is adjusted. This enables users to obtain an out-of-screen or in-screen effect that better meets their individual needs, thereby improving the interactivity and viewing experience of naked-eye 3D display.

[0039] In some embodiments, step S102 is an optional step, meaning that steps S101, S103 to S104 can be implemented independently to constitute a separate embodiment. In this embodiment, step S104, "updating the highlighted focus area to the focus area corresponding to the target zero-parallax plane," can be replaced by: highlighting the focus area corresponding to the target zero-parallax plane. In this embodiment, without needing to first determine and highlight the focus area corresponding to the reference zero-parallax plane, the user can directly set the position of the zero-parallax plane (e.g., drag or click) to adjust the zero-parallax plane.

[0040] Considering that directly using the scene bounding box center, simple arithmetic mean, or a small number of outliers to determine the reference zero-parallax plane may lead to deviations from the main effective depth range due to depth distribution skew and outlier spatial primitives, some embodiments, such as..., are used to improve the accuracy of determining the reference zero-parallax plane. Figure 2As shown, in step S101 above, determining the reference zero disparity plane of the 3D scene model may include: S201: Obtain the camera space depth of each spatial primitive in the 3D scene model.

[0041] The camera spatial depth of a spatial primitive can be the axial projection component of the center position of the spatial primitive in the forward direction of the camera, that is, the Z coordinate component of the center position of the spatial primitive in the camera coordinate system.

[0042] For example, for a spatial primitive j, the world coordinates p of the spatial primitive j are calculated based on the center coordinates of the spatial primitive j and the world transformation matrix of the three-dimensional scene entity. j Let the position coordinates of the observing camera be c, and the forward unit vector of the camera be f. Then the camera spatial depth z of the spatial primitive j is... j Calculate using the following formula: ; where the operator " "" indicates vector dot product. Here, the observation camera position and the camera's forward unit vector can be included in the camera parameters used when rendering the current 3D scene.

[0043] S202: Sort the camera spatial depth of each spatial primitive, and based on the sorting result, take the depth corresponding to the first preset quantile as the effective near depth and the depth corresponding to the second preset quantile as the effective far depth, wherein the first preset quantile is less than the second preset quantile.

[0044] Therefore, the effective near depth and effective far depth obtained based on the quantile statistics method can reflect the depth range of the main content of the scene. This eliminates outliers that are too far or too close, and can reduce the interference of individual distant sky or nearby occluders on the reference zero parallax plane.

[0045] In some embodiments, to more accurately determine the reference zero-parallax plane, before sorting the camera spatial depths of each spatial primitive, spatial primitives with invalid depth values ​​and those located in front of the near clipping plane can be filtered out based on their camera spatial depths. For example, spatial primitives with empty or non-finite depth values ​​(such as infinity) are considered invalid and filtered out. Spatial primitives located in front of the near clipping plane are also filtered out because their depth values ​​may be negative or less than the distance to the near clipping plane, as they are not within the effective visibility range. This avoids interference from invalid data or samples in front of the near clipping plane, allowing the effective near depth and effective far depth to more accurately reflect the depth distribution of the main content of the scene, thereby more accurately determining the reference zero-parallax plane.

[0046] The first and second preset quantiles can be set according to the actual application. For example, the first preset quantile can be any quantile from the 1st to the 5th percentile, and the second preset quantile can be any quantile from the 95th to the 99th percentile. For instance, using the 1st percentile as the first preset quantile and the 99th percentile as the second preset quantile means sorting the depths and obtaining the 1st and 99th percentiles; the depth corresponding to the 1st percentile is the effective near depth. The depth corresponding to the 99th percentile is the effective far depth. .

[0047] In some examples, the quantile positions are used Determine, where L is the number of valid samples, q is the quantile, and round indicates rounding to the nearest integer. q can take values ​​of 0.01, 0.50, and 0.99. This yields z. n =Q 0.01 (z), z med =Q 0.50 (z), z f =Q 0.99 (z). Wherein, z n For effective near depth, z med For the center depth (i.e., the midpoint depth), z f For effective long-distance depth.

[0048] In some embodiments, if the 3D scene model contains a very large number of spatial primitives, such as exceeding one million, directly sorting the camera spatial depth of all spatial primitives would incur significant computational overhead. To control statistical time and memory usage, equally spaced sampling can be performed before sorting. Specifically, the equally spaced sampling step size can be set according to the total number of spatial primitives, ensuring that the number of spatial primitives participating in the sorting does not exceed a preset number. For example, if the preset number M is 200,000 and the number of spatial primitives is Ns, the equally spaced sampling step size can be determined according to the following calculation method. .in, By traversing the spatial primitives with equally spaced sampling steps, the effective depth array is obtained and sorted in ascending order. This greatly reduces the computational load and quickly determines the reference zero-disparity plane.

[0049] S203: Determine the midpoint between the effective near depth and the effective far depth in the inverse depth domain, and set the reference zero parallax plane at the depth corresponding to the midpoint.

[0050] After obtaining the effective near depth and effective far depth, calculate the reciprocals of the effective near depth and effective far depth respectively to obtain the first inverse depth value and the second inverse depth value; calculate the average of the first inverse depth value and the second inverse depth value to obtain the midpoint of the inverse depth domain; convert the midpoint of the inverse depth domain into a depth value to obtain the reference zero disparity plane depth.

[0051] For example, the depth of the reference zero-parallax plane can be calculated using the following formula: ;in, Depth of the reference zero parallax plane, For effective near depth, This is the effective far depth. This value method makes the effective near depth and effective far depth relative to each other. The reciprocal depth deviations are equal in magnitude, that is: Due to the lateral pixel parallax of off-axis cameras and Proportional to this value, under the same camera spacing, points located at the effective near depth and points located at the effective far depth will produce physical pixel parallax of equal size and opposite direction, which can balance the out-of-screen range and the in-screen range.

[0052] In this embodiment, by taking the midpoint between the effective near depth and the effective far depth in the inverse depth domain and setting the reference zero disparity plane at the depth corresponding to the midpoint, the reference zero disparity plane can more reasonably represent the visual center of the scene subject in terms of disparity distribution, while avoiding the deviation caused by directly averaging the depth.

[0053] It should be noted that steps S201 to S203 described above can be implemented individually to constitute an independent embodiment, or they can be combined with... Figure 1 Some of the steps shown constitute an independent embodiment. For example, steps S201 to S203 can constitute an independent embodiment with step S102.

[0054] In some embodiments, such as Figure 3 As shown, in step S102 above, determining the focus area corresponding to the reference zero parallax plane includes: S301: Determine the focal plane distance corresponding to each spatial primitive based on the depth difference between each spatial primitive in the 3D scene model and the reference zero parallax plane.

[0055] For example, for each spatial primitive in a 3D scene model, the focal plane distance corresponding to that spatial primitive can be obtained based on the absolute value of the depth difference between the depth value of that spatial primitive and the depth of the reference zero-parallax plane. The focal plane distance reflects the degree to which the spatial primitive deviates from the reference zero-parallax plane. The smaller the focal plane distance, the closer the spatial primitive is to the reference zero-parallax plane; the larger the focal plane distance, the farther the spatial primitive is from the reference zero-parallax plane.

[0056] S302: For each spatial primitive, calculate the peak focus weight based on the focal plane distance and display bandwidth corresponding to the spatial primitive, wherein the display bandwidth is used to limit the width of the focus area in the depth direction.

[0057] S303: Based on spatial primitives with a peak focus weight greater than zero, the focus area corresponding to the reference zero parallax plane is determined. When the focal plane distance corresponding to the spatial primitive gradually increases from a preset proportion of the display bandwidth to the display bandwidth, the peak focus weight gradually decreases from the maximum weight value to zero.

[0058] In this embodiment, the relationship between the peak focus weight and the focal plane distance can be as follows: when the focal plane distance is less than a preset ratio of the display bandwidth, the peak focus weight is the preset maximum weight value; when the focal plane distance gradually increases from the preset ratio to the display bandwidth, the peak focus weight gradually decreases; when the focal plane distance reaches the display bandwidth, the peak focus weight is zero.

[0059] The preset ratio can be set according to the actual application. For example, the preset ratio is 0.35, which means that when the focal distance is less than 0.35 times the display bandwidth, the peak focus weight is the maximum weight value. As the focal distance gradually increases from 0.35 times the display bandwidth to the display bandwidth, the weight gradually decreases from the maximum weight value to 0 through a smooth step function. When the focal distance reaches or exceeds the display bandwidth, the weight is 0. In this way, the peak focus weight forms a continuous weight distribution near the zero parallax plane.

[0060] In practical implementation, the peak focus weight can be directly calculated during the vertex shading stage of the 3D Gaussian sputtered material. For the spatial primitive j, let its depth be... The current depth of the zero parallax plane is The display bandwidth is B. First, calculate the focal distance of this spatial primitive from the zero-parallax plane. Then, the peak focus weight of spatial primitive j is calculated using a smooth step function. For example, the smooth step function is: Where E is the activation coefficient, which controls whether peak focus display is enabled for the current spatial primitive. Its default value is 1, indicating that it is enabled. S is the peak intensity, which can be set to a preset value in the range of (0, 1). For example, S=1, S can be used to adjust the maximum mixing intensity of the highlight color.

[0061] In the above smoothing step function, when the focal plane distance is less than 0.35B, smoothstep returns to 0, and the peak focus weight is the maximum weight value (i.e., When the focal distance increases from 0.35B to B, the smoothstep return value smoothly transitions from 0 to 1, and the peak focus weight correspondingly gradually decreases from its maximum weight value to zero; when the focal distance reaches or exceeds B, the peak focus weight is zero. In this way, the peak focus weight of the spatial primitive forms a continuously changing weight distribution near the zero parallax plane. It can be understood that 0.35B represents a preset ratio of 0.35 for the display bandwidth B, and this preset ratio can be set to a positive number not greater than 0.5.

[0062] In this embodiment, a peak focus weight is calculated for each spatial primitive based on the focal plane distance and display bandwidth corresponding to each spatial primitive. Spatial primitives are then selected based on the weight to determine the focus area corresponding to the reference zero parallax plane. Since the peak focus weight changes continuously, the boundary of the focus area can be smoothed, the high-brightness display effect is more natural, and the user can intuitively perceive the position of the zero parallax plane.

[0063] It should be noted that steps S301 to S303 described above can be implemented individually to constitute an independent embodiment, or they can be combined with... Figure 1 Some of the steps shown constitute independent embodiments; for example, step S101 and steps S301 to S303 constitute independent embodiments; or Figure 2 All the steps shown can constitute an independent embodiment with steps S301 to S303.

[0064] In some embodiments, step S102 above, highlighting the focus area corresponding to the reference zero parallax plane, includes: For each spatial primitive within the focus area corresponding to the reference zero parallax plane, the original color of the spatial primitive is mixed with the first preset color according to the peak focus weight of the spatial primitive to obtain the preview color of the spatial primitive, and the spatial primitive is displayed in the preview color.

[0065] Exemplarily, exemplaryly, in obtaining the peak focus weight of spatial primitive j Then, the colors of the spatial primitive j can be mixed in the following way: Among them, C j The original color of the spatial primitive j. These are the preview colors used for display after mixing; The first preset color is (e.g., green), which can be set to (0.12, 1.00, 0.48); `mix` is a linear blending function, representing the blending time when... When it is 0, Equals C j ,when When the weight is the maximum value (e.g., 1), The first preset color is equal to the first color; for other cases, a weighted sum of the two is calculated. When As the maximum weight value gradually decreases to 0, the preview color gradually transitions to the original color until it equals the original color.

[0066] In this way, by mixing the original color of the spatial primitive with a first preset color according to the peak focus weight, a continuously gradient highlight display can be formed in the focus area. This display method preserves the original texture and color details of the spatial primitive, while highlighting the area near the zero parallax plane, improving the user's perception accuracy of the zero parallax plane position and the intuitiveness of adjustment interaction.

[0067] In some embodiments, such as Figure 4 As shown, the display bandwidth is determined in the following ways, including: S401: Multiply the difference between the effective near depth and effective far depth of the 3D scene model, the scene radius, and the depth of the reference zero parallax plane by their respective first preset multiples, and use the maximum value of the resulting product as the base bandwidth.

[0068] The effective near depth and effective far depth can be obtained by sorting the spatial primitive depths and taking the quantiles as described above, which will not be repeated here.

[0069] The scene radius can be the radius of the sphere or bounding box that encloses the 3D scene model, or it can be determined based on the maximum distance from the spatial primitives in the 3D scene to the center of the scene.

[0070] For the three quantities—the difference between effective near depth and effective far depth, the scene radius, and the depth of the reference zero-parallax plane—different first preset multiples can be set. Preferably, the first preset multiple corresponding to the difference between effective near depth and effective far depth is 0.018, the first preset multiple corresponding to the scene radius is 0.006, and the first preset multiple corresponding to the depth of the reference zero-parallax plane is 0.0015. The three products are calculated, and the maximum value among them is taken as the base bandwidth.

[0071] S402: The first value is the base bandwidth, the second value is the product of the depth of the reference zero parallax plane and the second preset multiple, and the third value is the product of the scene radius and the third preset multiple. The maximum value among the first, second and third values ​​is taken as the display bandwidth.

[0072] In actual use of peak focus display, to avoid a mathematically thin focal plane falling precisely between sparse three-dimensional Gaussian sputtering layers, resulting in insufficient spatial primitives being highlighted within the focus area, in this embodiment, the display bandwidth is further constrained to a second preset multiple based on the base bandwidth, the depth of the reference zero parallax plane, and a third preset multiple of the scene radius. The second and third preset multiples can be set separately.

[0073] For example, Where B is the final display bandwidth used. Based on bandwidth, Depth of the reference zero parallax plane, The radius is the scene radius.

[0074] Thus, by combining the effective depth range of the 3D scene, the scene radius, and the depth of the reference zero-parallax plane to jointly determine the final display bandwidth, it is possible to ensure that the display bandwidth has an appropriate width. Compared with using a uniform and fixed display bandwidth in different 3D scenes, this embodiment can reduce visual discomfort caused by excessive bandwidth, and can also more accurately determine the position of the zero-parallax plane, or reduce the situation where the user has difficulty intuitively perceiving the position of the zero-parallax plane due to insufficient bandwidth. In addition, even if the depth distribution of the main scene is narrow or the spatial primitives near the zero-parallax plane are sparse, the focus area will not be too thin, thereby ensuring that the peak focus highlight display is visually continuous and improving the intuitiveness of the user's operation when adjusting the zero-parallax plane.

[0075] In some embodiments, such as Figure 5 As shown, in step S103 above, in response to the user's adjustment operation on the position of the zero parallax plane, the target zero parallax plane is determined based on the adjustment operation and the reference zero parallax plane, including: S501: Based on the target position set by the adjustment operation and the position of the reference zero parallax plane, determine the corresponding target physical pixel offset, wherein the target physical pixel offset is the horizontal offset in units of physical pixels of the display panel.

[0076] For example, the corresponding target physical pixel offset can be determined based on a pre-established mapping relationship between depth distance and physical pixel offset. Depth distance is the distance in the depth direction between the target position and the location of the reference zero-parallax plane (which can be represented by a depth difference). When the target position is behind the reference zero-parallax plane in the depth direction (further from the viewing camera), the target physical pixel offset is positive, indicating that the scene content tends to be off-screen; when the target position is in front of the reference zero-parallax plane in the depth direction (closer to the viewing camera), the target physical pixel offset is negative, indicating that the scene content tends to be in-screen. The absolute value of the target physical pixel offset increases as the distance between the target position and the reference zero-parallax plane increases.

[0077] S502: Based on the target physical pixel offset, translate each camera in the virtual camera array laterally and adjust the lateral principal point of each camera projection matrix to set the target zero parallax plane at the target position, and the contents in front of and behind the target zero parallax plane form lateral parallax in opposite directions.

[0078] A virtual camera array consists of multiple virtual cameras, such as 30 virtual cameras. These virtual cameras are arranged at equal intervals and maintain the same rotational orientation, without using convergent rotation (i.e., toe-in rotation, also known as "inward tilt rotation"). Each camera in the virtual camera array has a preset lateral translation amount relative to the center of the virtual camera array.

[0079] Once the target physical pixel offset is determined, the base camera spacing can be recalculated based on this offset, resulting in a new lateral translation for each camera. After obtaining the new lateral translation for each camera, while keeping the camera rotation constant, each camera is translated laterally along the right vector of the base camera. Simultaneously, the lateral principal point offset of each camera's projection matrix is ​​modified. For example, for each camera, the parameters corresponding to the lateral principal point in the projection matrix can be adjusted based on the camera's lateral translation and the depth of the target zero-parallax plane. This ensures that the lateral projection positions of the spatial primitives located on the target zero-parallax plane remain consistent across multiple viewpoints, thereby creating zero parallax at the target zero-parallax plane.

[0080] As an example, an asymmetric projection can be formed by modifying the lateral principal point terms of the perspective projection matrix, for example, as follows: ;in, The horizontal principal point offset of the projection matrix before adjustment; The horizontal principal point offset of the adjusted projection matrix; These are preset coefficients related to the focal length in the projection matrix; This represents the lateral translation of the camera with index number i relative to the center of the virtual camera array; The depth of the reference zero-parallax plane.

[0081] In this embodiment, since all virtual cameras maintain the same rotational posture, and parallax control is achieved only through lateral translation and adjustment of the lateral principal point term of the projection matrix, spatial primitives located in front of the target zero-parallax plane will shift in one direction between adjacent virtual camera viewpoints, forming a first-direction lateral parallax; spatial primitives located behind the target zero-parallax plane will shift in another direction, forming a second-direction lateral parallax. The first and second directions are opposite, thus producing natural out-of-screen and in-screen effects.

[0082] It should be noted that steps S501 to S502 described above can be implemented individually to constitute an independent embodiment, or they can be combined with... Figure 1 Some of the steps shown constitute independent embodiments; for example, step S101 and steps S501 to S502 constitute independent embodiments; or Figure 2 All the steps shown can constitute an independent embodiment with steps S501 to S502.

[0083] In some embodiments, determining the corresponding target physical pixel offset based on the target position set by the adjustment operation in step S501 above may include: Determine the depth offset of the target position set in the adjustment operation relative to the location of the reference zero parallax plane; Determine the target depth level into which the depth offset falls among multiple preset depth levels; According to the preset mapping relationship, the target physical pixel offset corresponding to the target depth level is determined. The preset mapping relationship includes the correspondence between depth level and physical pixel offset.

[0084] In this embodiment, multiple depth levels can be set to control the pixel parallax and camera spacing of the display panel. For example, 100 depth levels are set, each corresponding to a physical pixel offset. The physical pixel offset is measured in physical pixels of the display panel; for instance, when the display panel is an 8K display panel, the physical pixel offset is measured in physical pixels of the 8K display panel. Pixel parallax refers to the physical pixel offset of corresponding points in the left and right eye images on the display panel. Unit camera spacing refers to the horizontal distance between two adjacent virtual cameras in the virtual camera array.

[0085] In some examples, 10 baseline target offset values ​​can be predetermined, such as 8, 12, 16, 20, 24, 30, 40, 52, 66, and 100 physical pixels. These 10 baseline values ​​are then extended to 100 levels using linear interpolation, resulting in the physical pixel offsets corresponding to each of the 100 depth levels. The default depth level can be set to 67.

[0086] It should be noted that the target physical pixel offset can be positive or negative. A positive value indicates that the target position is behind the reference zero parallax plane; a negative value indicates that the target position is in front of the reference zero parallax plane. The depth level is positively correlated with the absolute value of the physical pixel offset, that is, the higher the depth level, the larger the absolute value of the physical pixel offset, and the farther the target position is from the reference zero parallax plane; conversely, the closer the target position is to the reference zero parallax plane, the lower the depth level.

[0087] When determining the target physical pixel offset, the corresponding physical pixel offset can be found according to the target depth level and a preset mapping relationship. The preset mapping relationship is, for example, the correspondence table between multiple depth levels and physical pixel offsets obtained through linear interpolation mentioned above.

[0088] In some embodiments, for content with different aspect ratios, proportional display and edge patching can be used, but the parallax calculation is still performed according to the physical output width of the display panel to ensure the consistency of the parallax amount in physical pixels. In this way, no matter how the displayed content is scaled or patched, the pixel offset corresponding to the adjustment of the zero parallax plane is calculated based on a uniform physical panel width, thus keeping the stereoscopic effect stable under different aspect ratios.

[0089] In some embodiments, in step S502 above, the step of horizontally translating each camera in the virtual camera array based on the target physical pixel offset includes: The pixel offset per unit camera spacing is calculated based on the physical output width of the display panel, the camera lateral projection coefficient, the maximum number of camera steps, the effective near depth of the 3D scene model, and the depth of the reference zero parallax plane. The base camera spacing is calculated based on the target physical pixel offset and the pixel offset generated by the unit camera spacing. For each camera in the virtual camera array, the difference between the camera index and the center camera index is multiplied by the base camera spacing to obtain the camera's lateral translation amount, and the camera is then laterally translated by this lateral translation amount.

[0090] For example, let the physical output width of the display panel be W, the horizontal projection factor of the camera be Px, and the maximum number of camera steps be m. max The effective near depth of the 3D scene model is The depth of the reference zero parallax plane is .in, This is the inverse depth value of the reference zero parallax plane. Effective near depth. It can be the effective near depth value obtained through quantile statistics as mentioned above.

[0091] The physical output width W refers to the number of physical pixels in the horizontal direction of the naked-eye 3D display panel, i.e., the horizontal output width in units of physical pixels. For example, if an 8K panel has 7680 physical pixels in the horizontal direction, then the physical output width W can be 7680. The camera horizontal projection coefficient Px is determined based on the virtual camera's field of view, the aspect ratio of the output image, and the field of view definition mode (horizontal or vertical). It is used to convert the horizontal offset in camera space into pixel offset. Here, the camera horizontal projection coefficient can be calculated using the basic perspective projection formula in existing technologies, which will not be elaborated further here.

[0092] Maximum camera steps (m) maxThis represents the maximum number of offset steps for any camera within the virtual camera array relative to the array center along the array's arrangement direction. Each discrete position on the array used to place a virtual camera is called a camera position; the number of steps represents the offset of the camera position relative to the center camera position. The maximum camera step count is used to convert unit camera spacing into pixel disparity at the array edge viewpoints; the pixel disparity at the array edge viewpoints is the horizontal pixel offset of the outermost viewpoint image relative to the center viewpoint image. In this embodiment, 30 viewpoints can be used. If the virtual camera array contains 30 virtual cameras, the camera index range is 0~29, the reference camera index (e.g., the center camera index) is 14, the index offset range is -14 to +15, and the maximum absolute index step count is 15, i.e., the maximum camera step count m. max =15.

[0093] Specifically, you can first calculate the pixel offset K caused by the unit camera spacing, for example, by calculating the pixel offset K using the following formula: ;

[0094] in, and These are the inverse depth values ​​corresponding to the effective near depth and the inverse depth values ​​of the reference zero parallax plane, respectively. Due to the lateral pixel parallax generated by the off-axis camera and... Proportional The pixel offset K is the inverse depth difference of the spatial primitive relative to the reference zero parallax plane. Therefore, the pixel offset K reflects the pixel offset that a unit camera spacing can cause at the near end under the current camera settings and the position of the zero parallax plane. That is, when the horizontal spacing between two adjacent virtual cameras is 1 unit length (e.g., 1 mm), the pixel offset generated at the effective near-end depth.

[0095] Then, based on the target physical pixel offset D t Given the pixel offset K, calculate the base camera spacing Δc. Target physical pixel offset D. t This characterizes the target physical pixel parallax corresponding to the current depth level, i.e., the physical pixel offset that the user expects to produce on the display panel by adjusting the zero parallax plane. For example, the target physical pixel offset D can be directly expressed as... t Divide by the pixel offset K to calculate the basic camera spacing Δc.

[0096] In some examples, the basic camera spacing Δc can be calculated using the following formula: ; Where ε is a very small positive number used to avoid division by zero when K is close to zero; r is the scene radius. This clamp operation is used to limit the base camera spacing Δc to between one ten-thousandth and one-quarter of the scene radius r, ensuring that the camera spacing is neither too small to cause imperceptible parallax changes, nor too large to cause stereoscopic image distortion.

[0097] After obtaining the base camera spacing Δc, for each camera in the virtual camera array, the difference between the camera's index and the center camera index is calculated, and this difference is multiplied by the base camera spacing to obtain the camera's lateral translation. For example, for the camera with index i, its lateral translation xi can be expressed as: Where c is the center camera index of the camera array. When the virtual camera array contains 30 cameras, the index range is 0~29, and the center camera index c=14. Therefore, the lateral translation of the camera with index 0 is -14Δc, and the lateral translation of the camera with index 29 is 15Δc. Subsequently, each camera translates according to its own lateral translation amount, and the lateral principal point offset of the corresponding projection matrix is ​​adjusted to form zero disparity at the target zero disparity plane. During the lateral translation process, each camera maintains the same rotational attitude.

[0098] In addition, the pixel offsets mentioned above are all physical panel pixels and do not change with the cropping or padding of the displayed content, thus ensuring that parallax control remains stable and consistent under different aspect ratio display conditions.

[0099] It is worth noting that, since the reference zero-parallax plane can be set at the depth corresponding to the midpoint between the effective near depth and effective far depth of the 3D scene model by calculating the midpoint in the inverse depth domain, this method makes the near-end and far-end inverse depths symmetrical about the reference zero-parallax plane. At the reference zero-parallax plane, the spatial point corresponding to the effective near depth is expected to generate approximately +D. t The physical pixel offset, the spatial point corresponding to the effective far depth is expected to generate approximately -D t The physical pixel offset causes the effective near point and the effective far point to produce physical pixel parallax of equal size but opposite direction at the same camera spacing.

[0100] When the target position set by the user's adjustment deviates from the reference zero-parallax plane, the inverse depth difference between the effective near depth and the effective far depth relative to the target zero-parallax plane is no longer equal. Therefore, the physical pixel offsets corresponding to spatial points at the effective near depth and at the effective far depth will also be different. At this time, by determining the camera spacing based on the target physical pixel offset, and then obtaining the lateral translation of each camera, the lateral translation of each camera in the virtual camera array can make the parallax at the target zero-parallax plane zero.

[0101] Through the above process, the target physical pixel offset set by the user adjustment operation is converted into the lateral translation of each camera in the virtual camera array, thereby changing the position of the zero parallax plane and forming out-of-screen and in-screen parallax in opposite directions, realizing flexible control of the naked-eye 3D display stereoscopic effect.

[0102] In some embodiments, the adjustment operation is a sliding operation applied to a preset control, and the sliding direction of the sliding operation corresponds to the forward and backward movement of the zero parallax plane in the depth direction.

[0103] For example, the preset control can be a slider, a lever, or other adjustable control. Taking a slider as an example, the slider has a sliding slider, and the user can adjust the position of the zero plane on the Z-axis (i.e., the depth direction) by dragging the slider.

[0104] Figure 6a This is a schematic diagram of the interface when the zero parallax plane is located in the center of the scene, as provided in an embodiment of this application. Figure 6b This is a schematic diagram of the interface when the zero parallax plane is located near the scene, as provided in an embodiment of this application. Figure 6c This is a schematic diagram of the interface when the zero parallax plane is located at a distance in the scene, as provided in an embodiment of this application. Figures 6a to 6c As shown, the interactive window of the display interface has a switch for displaying the focal plane of the point cloud. When the switch is on, while displaying the 3D scene model, only the focus area corresponding to the zero parallax plane of the target is highlighted in green; the on-screen area and the in-screen area are not highlighted. The preset control can be a horizontal slider in the interactive window on the right side of the display interface, with a slider on the horizontal slider to adjust the focal plane distance (i.e., changing the position of the zero parallax plane by changing the focal plane distance). The left end of the horizontal slider corresponds to the near end of the scene, and the right end corresponds to the far end of the scene. When the user moves the slider to near the left end, the zero parallax plane is closer to the effective near depth, and the focus area is highlighted in the first preset color (represented in green in the figure) and concentrated in the near part of the scene; when the slider is in the middle position, the zero parallax plane is in the middle of the main body of the scene, and the highlighted area (represented in green in the figure) is correspondingly in the middle of the scene; when the slider is moved to near the right end, the zero parallax plane is closer to the effective far depth, and the highlighted area (represented in green in the figure) is concentrated in the far part of the scene.

[0105] In this embodiment, the user can adjust the zero-parallax plane to the desired position through adjustment operations. For example, if the user wants the entire scene to be behind the screen (i.e., completely in-screen), the zero-parallax plane can be moved forward so that all spatial primitives are behind the zero-parallax plane; if the user wants all content to protrude from the screen (i.e., completely out-of-screen), the zero-parallax plane can be moved backward so that all spatial primitives are in front of the zero-parallax plane; if the user wants some content to be out of screen and some content to be in-screen, the zero-parallax plane can be set in the middle of the scene.

[0106] In some embodiments, the method further includes: when highlighting the focus area corresponding to the target zero parallax plane, highlighting the on-screen area with a second preset color, and / or, highlighting the in-screen area with a third preset color.

[0107] In this embodiment, the system can partition and color spatial primitives according to the expected physical pixel disparity. Let the basic camera spacing be Δc, the physical output width of the display panel be W, the camera lateral projection coefficient be Px, and the maximum camera step size be 15. Then, the pixel disparity D(z) corresponding to the spatial primitive located at depth z can be expressed as: ; in, Let D(z) be the inverse depth value of the reference zero-parallax plane, and 1 / z be the inverse depth value of the spatial primitive. In this formula, when D(z) is positive, it means that the spatial primitive is closer to the viewer relative to the target zero-parallax plane, which can create an out-of-screen effect; when D(z) is negative, it means that the spatial primitive is farther away from the viewer relative to the target zero-parallax plane, which can create an in-screen effect; when D(z) is zero, it means that the center of the spatial primitive is located on the target zero-parallax plane.

[0108] In some examples, when highlighting, areas with positive D(z) can be displayed in orange-red, and areas with negative D(z) in blue-purple; when the absolute value of D(z) is close to zero, it is displayed in green. For example, the interface content can be judged with a screen surface tolerance of ±1 8K physical pixels, and the shading transition can be set between approximately 0.5 and 1.25 pixels.

[0109] In this way, users can intuitively observe the out-of-screen area, the in-screen area, and the area near the zero parallax plane in the scene through the color distribution.

[0110] Figure 6d This is one of the schematic diagrams of highlight color distribution in a three-dimensional scene provided in the embodiments of this application. Figure 6e This is the second schematic diagram of the distribution of highlight colors in a three-dimensional scene provided in the embodiments of this application. Figure 6f This is the third schematic diagram of the distribution of highlight colors in a three-dimensional scene provided in this application embodiment. For example... Figures 6d to 6fAs shown, when the switch for displaying the focal plane of the point cloud within the interactive window of the display interface is off, while displaying the 3D scene model, the outgoing area is shown in orange-red and the incoming area in blue-purple. The display interface can simultaneously highlight the focus area of ​​the zero parallax plane in green, display the outgoing area in orange-red, and display the incoming area in blue-purple. Specifically, in... Figure 6d In the middle, the slider is positioned near the center of the horizontal slider bar, with the green highlighted area near the center of the scene, and the orange-red and blue-purple areas located in front of and behind the green area respectively in the depth direction of the scene; Figure 6e In the middle, as the slider moves to the left on the horizontal slider bar, the zero parallax plane moves forward synchronously, the green highlight focus area moves forward accordingly, the orange-red out-of-screen area shrinks, and the blue-purple in-screen area expands. Figure 6f In the middle, the slider moves to the right on the horizontal slider bar, the zero parallax plane moves backward in sync, the green highlight focus area moves backward accordingly, the orange-red out-of-screen area expands, and the blue-purple in-screen area shrinks until it becomes invisible.

[0111] It should be noted that this shading display is only used to assist users in judging and adjusting the position of the zero parallax plane, and will not change the actual depth information of the 3D scene or affect the final stereoscopic rendering result.

[0112] In some embodiments, after adjusting the zero-parallax plane to the target position to obtain the target zero-parallax plane, multiple viewpoint images can be generated based on the target zero-parallax plane and color sub-pixel interleaving can be performed to output the interleaved image to the naked-eye 3D terminal.

[0113] Specifically, after adjusting the projection matrices of each camera to set the zero-parallax plane at the target position, the virtual camera array rendering produces multiple base viewpoint images. These base viewpoint images are then interleaved according to a round-trip phase sequence, and an interleaved image is generated by combining phase parameters and RGB color sub-pixel selection rules. The round-trip phase sequence is a sequence of base viewpoint indices arranged in a first direction to the maximum index and then in a second direction to the minimum index. Phase parameters may include the raster period (Pitch), tilt angle tangent (Tan), and phase offset (Offset). The RGB color sub-pixel selection rules are used to determine the base viewpoint index corresponding to each color channel in the output pixel. The interleaved image is output to the display panel, where it is processed by a directional optics module to form multiple hardware optical viewpoints.

[0114] For example, the terminal actually has 60 hardware optical viewpoints. To reduce rendering and caching overhead, the software only generates 30 basic camera viewpoints and constructs a round-trip phase sequence of length 2N-1 under the default alternation order. When N=30, the sequence is: 0,1,…,28,29,28,…,1,0 (a total of 59 items). The display panel is, for example, a 7680×4320 display panel; that is, the interlacing map is output to the 7680×4320 display panel, and the 60 hardware optical viewpoints are formed through the directional optics module.

[0115] In some examples, for the normalized phase bin index k∈[0,58], the base viewpoint index q(k) is: q(k)=k, k<30; q(k)=58-k, 30≤k≤58; the above 59-item sequence is formed based on the round-trip multiplexing rule of 30 rendering source viewpoints in a single optical cycle, and does not represent the number of hardware viewpoints. The 60 hardware optical viewpoints are formed based on Pitch, Tan, Offset, output pixel coordinates, color subpixel phase, and directional optical structure.

[0116] Regarding RGB subpixel interleaving, for the output pixel coordinates (x, y), the output image is flipped in the Y direction to the panel row coordinates, and the base phase of the RGB color subpixels can be calculated with a coordinate step size of 3. The calculation formula is: base = Offset + 3y·Tan + 3(W-1-x); where W is the physical output width of the display panel. The red, green, and blue color channels are calculated with base+2, base+1, and base, respectively, to obtain the viewpoint index; the phase is first normalized by performing a positive modulus operation on the pitch, and then quantized to a sequence of 30 or 59 terms. Finally, the R, G, and B channels of the pixel are sampled from the base viewpoint image corresponding to their respective indices, i.e., R = view[q(base+2)].R; G = view[q(base+1)].G; B = view[q(base)].B.

[0117] Regarding the isolation between the peak display and the final interlaced map, before the final 8K export begins, the system saves the peak focus and in / out screen area display states, then disables both auxiliary shading methods, renders and interlaces 30 basic viewpoints; after the export is complete or fails and the camera is restored, the original display states are restored. Therefore, the peak green, in / out screen colors, and focal plane markers only exist in the debug preview and are not retained in the final interlaced map.

[0118] The various embodiments or implementation methods described in this specification are presented in a progressive manner. Each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other.

[0119] It is worth noting that, without contradiction, the foregoing method embodiments can be combined arbitrarily. For example, some or all of the steps of different embodiments can be combined arbitrarily, and one embodiment can be combined arbitrarily with the optional implementations of other embodiments.

[0120] Figure 7 This is a schematic diagram of the zero parallax plane adjustment device provided in an embodiment of this application. Figure 7 As shown, the zero parallax plane adjustment device 100 includes: The first determining module 101 is used to determine the reference zero disparity plane of the three-dimensional scene model and to determine the focus area corresponding to the reference zero disparity plane. High-brightness display module 102 is used to highlight the focus area corresponding to the reference zero parallax plane; The zero parallax adjustment module 103 is used to respond to the user's adjustment operation on the position of the zero parallax plane, and determine the target zero parallax plane based on the adjustment operation and the reference zero parallax plane; The second determining module 104 is used to determine the focusing area corresponding to the target zero parallax plane; The display module 105 is updated to update the highlighted focus area to the focus area corresponding to the target zero parallax plane.

[0121] In some embodiments, the first determining module 101 is configured to: Obtain the camera space depth of each spatial primitive in the three-dimensional scene model; The camera spatial depths of each spatial primitive are sorted, and based on the sorting results, the depth corresponding to the first preset quantile is taken as the effective near depth, and the depth corresponding to the second preset quantile is taken as the effective far depth, wherein the first preset quantile is less than the second preset quantile. In the inverse depth domain, the midpoint between the effective near depth and the effective far depth is determined, and the reference zero parallax plane is set at the depth corresponding to the midpoint.

[0122] In some embodiments, the first determining module 101 is configured to: Based on the depth difference between each spatial primitive in the three-dimensional scene model and the reference zero disparity plane, the focal plane distance corresponding to each spatial primitive is determined; For each of the spatial primitives, a peak focus weight is calculated based on the focal plane distance and display bandwidth corresponding to the spatial primitive, wherein the display bandwidth is used to limit the display range of the focus area along the depth direction; Based on the spatial primitive with a peak focus weight greater than zero, the focus area corresponding to the reference zero parallax plane is determined; wherein, when the focal plane distance corresponding to the spatial primitive gradually increases from a preset proportion of the display bandwidth to the display bandwidth, the peak focus weight gradually decreases from the maximum weight value to zero.

[0123] In some embodiments, the high-brightness display module 102 is used for: For each spatial primitive within the focus area corresponding to the reference zero parallax plane, the original color of the spatial primitive is mixed with a first preset color according to the peak focus weight of the spatial primitive to obtain the preview color of the spatial primitive, and the spatial primitive is displayed in the preview color.

[0124] In some embodiments, the apparatus further includes: a third determining module; the third determining module is configured to: The difference between the effective near depth and the effective far depth of the 3D scene model, the scene radius, and the depth of the reference zero parallax plane are multiplied by their respective first preset multiples, and the maximum value of the resulting product is used as the base bandwidth. The basic bandwidth is used as the first value, the product of the depth of the reference zero parallax plane and the second preset multiple is used as the second value, the product of the scene radius and the third preset multiple is used as the third value, and the maximum value among the first value, the second value and the third value is taken as the display bandwidth.

[0125] In some embodiments, the zero parallax adjustment module 103 is used for: Based on the target position set by the adjustment operation and the position of the reference zero parallax plane, the corresponding target physical pixel offset is determined. The target physical pixel offset is the horizontal offset in units of physical pixels of the display panel. Based on the target physical pixel offset, each camera in the virtual camera array is horizontally translated, and the horizontal principal point item of the projection matrix of each camera is adjusted to set the target zero parallax plane at the target position, and the contents in front of and behind the target zero parallax plane form a horizontal parallax with opposite directions.

[0126] In some embodiments, the zero parallax adjustment module 103 is specifically used for: Determine the depth offset of the target position set by the adjustment operation relative to the position of the reference zero disparity plane, wherein the reference position is the position of the effective near depth of the three-dimensional scene model in the inverse depth domain; Determine the target depth level into which the depth offset falls among multiple preset depth levels; According to a preset mapping relationship, the target physical pixel offset corresponding to the target depth level is determined, wherein the preset mapping relationship includes the correspondence between depth level and physical pixel offset.

[0127] In some embodiments, the zero parallax adjustment module 103 is specifically used for: The pixel offset per unit camera spacing is calculated based on the physical output width of the display panel, the camera lateral projection coefficient, the maximum number of camera steps, the effective near depth of the 3D scene model, and the depth of the reference zero parallax plane. The base camera spacing is calculated based on the target physical pixel offset and the pixel offset; For each camera in the virtual camera array, the difference between the camera's index and the center camera's index is multiplied by the base camera spacing to obtain the camera's lateral translation amount, and the camera is then laterally translated by the lateral translation amount.

[0128] In some embodiments, the adjustment operation is a sliding operation applied to a preset control, and the sliding direction of the sliding operation corresponds to the forward and backward movement of the zero parallax plane in the depth direction.

[0129] In some embodiments, the high-brightness display module 102 is further configured to: When highlighting the focus area corresponding to the target zero parallax plane, the on-screen area is highlighted with a second preset color, and / or the in-screen area is highlighted with a third preset color.

[0130] The zero-parallax plane adjustment device provided in this application belongs to the same concept as the zero-parallax plane adjustment method provided in the above embodiments. It can execute the zero-parallax plane adjustment method provided in any of the above embodiments and has the corresponding functional modules and beneficial effects for executing the zero-parallax plane adjustment method. Technical details not described in detail in this embodiment can be found in the specific content of the zero-parallax plane adjustment method provided in the above embodiments, and will not be repeated here.

[0131] It should be understood that the various modules in the zero parallax plane adjustment device can be implemented in the form of processor calling software, or the various modules can be implemented in the form of hardware circuits. The functions of some or all modules can be realized through the design of the hardware circuits, which can be understood as one or more processors.

[0132] This application also provides an electronic device, including a processor, a memory, and an executable program stored in the memory and executable by the processor. When the processor runs the executable program, it performs the steps of the zero parallax plane adjustment method provided in any of the foregoing embodiments.

[0133] This application also provides a storage medium storing an executable program thereon, which, when executed by a processor, implements the steps of the zero parallax plane adjustment method provided in any of the foregoing embodiments.

[0134] For ease of understanding, the following focuses on explaining the terminology used in this embodiment: In this application embodiment, the processor is a circuit with signal processing capabilities. In one implementation, the processor can be a circuit with instruction read and execute capabilities, such as a Central Processing Unit (CPU), a microprocessor, a Graphics Processing Unit (GPU) (which can be understood as a type of microprocessor), or a Digital Signal Processor (DSP). In another implementation, the processor can implement certain functions through the logical relationships of hardware circuits. The logical relationships of the aforementioned hardware circuits are fixed or reconstructable. For example, the processor is a hardware circuit implemented using an Application-Specific Integrated Circuit (ASIC) or a Programmable Logic Device (PLD), such as an FPGA. In a reconstructable hardware circuit, the processor loads a configuration document, implementing a cyclical process of hardware circuit configuration. This can be understood as the processor loading instructions to implement the functions of some or all of the above units or modules in a cyclical process. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a Neural Network Processing Unit (NPU), a Tensor Processing Unit (TPU), a Deep Learning Processing Unit (DPU), etc.

[0135] The computer-readable storage medium provided in this embodiment can execute the zero parallax plane adjustment method of the above embodiment. Its implementation principle and technical effect are similar to those of the above embodiment, and will not be repeated here.

[0136] The aforementioned computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0137] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the processor and the readable storage medium can exist as discrete components in an electronic device or a host device.

[0138] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0139] The various embodiments or implementation methods described in this specification are presented in a progressive manner. Each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other.

[0140] In the description of this specification, references to "one embodiment," "some embodiments," "illustrative embodiment," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with an embodiment or example is included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0141] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A zero-parallax plane adjustment method, characterized in that, The method includes: Determine the reference zero-parallax plane for the 3D scene model; Determine the focus area corresponding to the reference zero parallax plane and highlight the focus area corresponding to the reference zero parallax plane; In response to the user's adjustment operation on the position of the zero parallax plane, the target zero parallax plane is determined based on the adjustment operation and the reference zero parallax plane, and the focus area corresponding to the target zero parallax plane is determined. The highlighted focus area is updated to the focus area corresponding to the target zero parallax plane.

2. The method according to claim 1, characterized in that, The determination of the reference zero disparity plane for the 3D scene model includes: Obtain the camera space depth of each spatial primitive in the three-dimensional scene model; The camera spatial depths of each spatial primitive are sorted, and based on the sorting results, the depth corresponding to the first preset quantile is taken as the effective near depth, and the depth corresponding to the second preset quantile is taken as the effective far depth, wherein the first preset quantile is less than the second preset quantile. In the inverse depth domain, the midpoint between the effective near depth and the effective far depth is determined, and the reference zero parallax plane is set at the depth corresponding to the midpoint.

3. The method according to claim 1, characterized in that, Determining the focus area corresponding to the reference zero parallax plane includes: Based on the depth difference between each spatial primitive in the three-dimensional scene model and the reference zero disparity plane, the focal plane distance corresponding to each spatial primitive is determined; For each of the spatial primitives, a peak focus weight is calculated based on the focal plane distance and display bandwidth corresponding to the spatial primitive, wherein the display bandwidth is used to limit the display range of the focus area along the depth direction; Based on the spatial primitives with a peak focus weight greater than zero, the focus area corresponding to the reference zero parallax plane is determined. Specifically, when the focal distance corresponding to the spatial primitive gradually increases from a preset proportion of the display bandwidth to the display bandwidth, the peak focus weight gradually decreases from the maximum weight value to zero.

4. The method according to claim 3, characterized in that, The highlighting of the focus area corresponding to the reference zero parallax plane includes: For each spatial primitive within the focus area corresponding to the reference zero parallax plane, the original color of the spatial primitive is mixed with a first preset color according to the peak focus weight of the spatial primitive to obtain the preview color of the spatial primitive, and the spatial primitive is displayed in the preview color.

5. The method according to claim 3, characterized in that, The display bandwidth is determined in the following way: The difference between the effective near depth and the effective far depth of the 3D scene model, the scene radius, and the depth of the reference zero parallax plane are multiplied by their respective first preset multiples, and the maximum value of the resulting product is used as the base bandwidth. The basic bandwidth is used as the first value, the product of the depth of the reference zero parallax plane and the second preset multiple is used as the second value, the product of the scene radius and the third preset multiple is used as the third value, and the maximum value among the first value, the second value and the third value is taken as the display bandwidth.

6. The method according to claim 1, characterized in that, The step of responding to a user's adjustment operation on the position of the zero-parallax plane, and determining the target zero-parallax plane based on the adjustment operation and the reference zero-parallax plane, includes: Based on the target position set by the adjustment operation and the position of the reference zero parallax plane, the corresponding target physical pixel offset is determined. The target physical pixel offset is the horizontal offset in units of physical pixels of the display panel. Based on the target physical pixel offset, each camera in the virtual camera array is horizontally translated, and the horizontal principal point item of the projection matrix of each camera is adjusted to set the target zero parallax plane at the target position, and the contents in front of and behind the target zero parallax plane form a horizontal parallax with opposite directions.

7. The method according to claim 6, characterized in that, The step of determining the corresponding target physical pixel offset based on the target position set by the adjustment operation and the position of the reference zero disparity plane includes: Determine the depth offset of the target position set by the adjustment operation relative to the location of the reference zero parallax plane; Determine the target depth level into which the depth offset falls among multiple preset depth levels; According to a preset mapping relationship, the target physical pixel offset corresponding to the target depth level is determined, wherein the preset mapping relationship includes the correspondence between depth level and physical pixel offset.

8. The method according to claim 6 or 7, characterized in that, The process of horizontally shifting each camera in the virtual camera array based on the target physical pixel offset includes: The pixel offset per unit camera spacing is calculated based on the physical output width of the display panel, the camera lateral projection coefficient, the maximum number of camera steps, the effective near depth of the 3D scene model, and the depth of the reference zero parallax plane. The base camera spacing is calculated based on the target physical pixel offset and the pixel offset generated by the unit camera spacing. For each camera in the virtual camera array, the difference between the camera's index and the center camera's index is multiplied by the base camera spacing to obtain the camera's lateral translation amount, and the camera is then laterally translated by the lateral translation amount.

9. The method according to any one of claims 1 to 7, characterized in that, The adjustment operation is a sliding operation applied to a preset control, and the sliding direction of the sliding operation corresponds to the forward and backward movement of the zero parallax plane in the depth direction.

10. The method according to claim 1, characterized in that, The method further includes: When highlighting the focus area corresponding to the target zero parallax plane, the on-screen area is highlighted with a second preset color, and / or the in-screen area is highlighted with a third preset color.

11. A zero parallax plane adjustment device, characterized in that, The device includes: The first determining module is used to determine the reference zero parallax plane of the three-dimensional scene model and to determine the focus area corresponding to the reference zero parallax plane. The high-brightness display module is used to highlight the focus area corresponding to the reference zero-parallax plane; The zero parallax adjustment module is used to respond to the user's adjustment operation on the position of the zero parallax plane, and determine the target zero parallax plane based on the adjustment operation and the reference zero parallax plane. The second determining module is used to determine the focus area corresponding to the target zero parallax plane; The display module is updated to update the highlighted focus area to the focus area corresponding to the target zero parallax plane.

12. An electronic device, characterized in that, It includes a processor, a memory, and an executable program stored in the memory and capable of being run by the processor, wherein the processor, when running the executable program, performs the zero parallax plane adjustment method as described in any one of claims 1 to 10.

13. A storage medium, characterized in that, It stores an executable program that, when executed by a processor, implements the zero parallax plane adjustment method as described in any one of claims 1 to 10.

Images