Dynamic projection interactive calibration method and system for annular science magic space

Abstract

The invention discloses a dynamic projection interaction calibration method and system for an annular science magic space, and belongs to the technical field of projection interaction. The method comprises the steps of obtaining three-dimensional point cloud data of an annular wall surface and spatial pose information of each projector, constructing a wall surface point cloud model, completing projector calibration according to the wall surface point cloud model, and obtaining a pose parameter data set; and based on the data set, mapping each frame of pixel of a preset picture sequence to a corresponding position of the wall surface in real time. And if user interaction is detected, determining an interaction area and executing corresponding logic. In the process, projection feedback images are continuously collected, dynamic calibration and tiny compensation are carried out in combination with pose parameters, calibration projection coordinates and real-time interaction response data are generated, and accordingly, all projectors are controlled to complete real-time and stable projection display on the annular wall surface. According to the scheme, the projection picture is highly consistent with the annular wall surface and user operation, dynamic calibration and real-time interaction are realized, the display precision and stability of projection are improved, and the immersion experience of a user is enhanced.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of projection interaction, and particularly relates to a dynamic projection interaction calibration method and system for a ring-shaped science fiction space. BACKGROUND

[0002] In an immersive experience scene, a ring-shaped wall surface projection technology can create a strong immersive atmosphere for users due to its unique space surrounding effect. However, to achieve high-quality projection display and accurate interaction, it is necessary to overcome the difficulties of wall surface three-dimensional modeling, projector calibration, accurate mapping of pictures, and real-time interaction, and it is necessary to develop a projection system that is efficient, accurate, and has dynamic calibration capability.

[0003] In the early projection technology, a projector with basic parameters is selected, a fixed static picture material is prepared by software, a simple support is built, and the projector is installed thereon. After roughly adjusting the position by manual experience, the picture is basically clear and covers part of the wall surface by manually adjusting the focus and angle. Then, a fixed projection mode is set to continuously display the picture. During the display, personnel are arranged to regularly check the operation of the projector and the display of the picture, and manual processing is performed when problems occur.

[0004] However, the early projection scheme has a fixed picture, which cannot be adjusted according to the changes in user behavior, resulting in a low interactive experience for users. In addition, when facing a complex wall surface, the picture is easily distorted and twisted, and the visual effect is greatly reduced. Moreover, the dynamic calibration function is missing, and slight changes in the projection device and the wall surface will cause the picture to be misaligned and blurred, and the stability is poor. SUMMARY

[0005] In order to overcome the above-mentioned defects, the present application is proposed to solve or at least partially solve the technical problems of the prior art, i.e., the early projection scheme has a fixed picture, which cannot be adjusted according to the changes in user behavior, resulting in a low interactive experience for users. In addition, when facing a complex wall surface, the picture is easily distorted and twisted, and the visual effect is greatly reduced. Moreover, the dynamic calibration function is missing, and slight changes in the projection device and the wall surface will cause the picture to be misaligned and blurred, and the stability is poor.

[0006] In a first aspect, the present application provides a dynamic projection interaction calibration method for a ring-shaped science fiction space, the method comprising: obtaining three-dimensional point cloud data of a ring-shaped wall surface in the ring-shaped science fiction space, and obtaining spatial pose information of each projector in the ring-shaped science fiction space, generating a ring-shaped wall surface point cloud model based on the three-dimensional point cloud data, and calibrating each projector based on the ring-shaped wall surface point cloud model and the spatial pose information to obtain a pose parameter data set of each projector; based on the pose parameter data set, controlling each projector to map the real-time state of each frame of pixels in a preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface; if user interaction data of a user on the ring-shaped wall surface is detected, an interaction area is determined based on the user interaction data and corresponding logic is executed; a projection feedback image of the ring-shaped wall surface is continuously acquired during execution of the corresponding logic, dynamic calibration and slight compensation processing are performed based on the projection feedback image and the pose parameter data set, calibration projection coordinates along the ring-shaped wall surface are generated, and real-time interaction response data corresponding to the calibration projection coordinates is generated; based on the calibration projection coordinate mapping and the real-time interaction response data, the projectors are controlled to perform projection display on the ring-shaped wall surface.

[0007] In a second aspect, the present application provides a dynamic projection interaction calibration system for a ring-shaped science fiction space, the system comprising: a calibration module configured to acquire three-dimensional point cloud data of a ring-shaped wall surface in the ring-shaped science fiction space, and acquire spatial pose information of the projectors in the ring-shaped science fiction space, generate a ring-shaped wall surface point cloud model based on the three-dimensional point cloud data, and calibrate the projectors based on the ring-shaped wall surface point cloud model and the spatial pose information to obtain a pose parameter data set of the projectors; a mapping module configured to control the projectors to map a real-time state of each frame of pixels in a preset picture sequence to corresponding projection coordinates on the ring-shaped wall surface based on the pose parameter data set; an interaction module configured to, if user interaction data of a user on the ring-shaped wall surface is detected, determine an interaction area based on the user interaction data and execute corresponding logic; a calibration module configured to continuously acquire a projection feedback image of the ring-shaped wall surface during execution of the corresponding logic, perform dynamic calibration and slight compensation processing based on the projection feedback image and the pose parameter data set, generate calibration projection coordinates along the ring-shaped wall surface, and generate real-time interaction response data corresponding to the calibration projection coordinates; a projection module configured to control the projectors to perform projection display on the ring-shaped wall surface based on the calibration projection coordinate mapping and the real-time interaction response data.

[0008] In a third aspect, an electronic device is provided, which comprises a processor, a memory, and a program or instructions stored on the memory and executable on the processor, the program or instructions being loaded and executed by the processor to perform the steps of the above-mentioned dynamic projection interaction calibration method for a ring-shaped science fiction space.

[0009] In a fourth aspect, a computer-readable storage medium is provided, which stores a plurality of program codes therein, the program codes being adapted to be loaded and executed by a processor to perform the steps of the above-mentioned dynamic projection interaction calibration method for a ring-shaped science fiction space.

[0010] The above-mentioned one or more technical solutions of the present application have at least one or more of the following beneficial effects: In the technical solutions of the present application, the projection picture is highly consistent with the ring-shaped wall surface and user operation, dynamic calibration and real-time interaction are realized, the display precision and stability of projection are improved, and the immersion experience of the user is enhanced. BRIEF DESCRIPTION OF DRAWINGS

[0011] The present application will become more fully understood from the detailed description given herein below and the accompanying drawings which are given by way of illustration only and thus are not intended to limit the present application. Like reference numerals refer to like elements throughout the drawings. Figure 1 is a main step flow diagram of a dynamic projection interactive calibration method of a ring-shaped science fiction space according to an embodiment of the present application; Figure 2 is a main step flow diagram of a dynamic projection interactive calibration method of a ring-shaped science fiction space according to an embodiment of the present application; Figure 3 is a main structure block diagram of a dynamic projection interactive calibration system of a ring-shaped science fiction space according to an embodiment of the present application; Figure 4 is a structure diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0012] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are some but not all of the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0013] Those skilled in the art can understand that although some embodiments herein include certain features but not other features included in other embodiments, the combination of features of different embodiments means to be within the scope of the present application and form different embodiments. For example, in the claims, any one of the claimed embodiments can be used in any combination.

[0014] Reference will now be made to the drawings, in which Figure 1 , Figure 1 is a main step flow diagram of a dynamic projection interactive calibration method of a ring-shaped science fiction space according to an embodiment of the present application. As shown in Figure 1 , the dynamic projection interactive calibration method of a ring-shaped science fiction space in the embodiments of the present application mainly includes the following steps S101-S105.

[0015] Step S101: Obtain the three-dimensional point cloud data of the annular wall surface in the annular science fiction space, and obtain the spatial pose information of each projector in the annular science fiction space, generate an annular wall surface point cloud model based on the three-dimensional point cloud data, and calibrate each projector based on the annular wall surface point cloud model and the spatial pose information to obtain a pose parameter data set of each projector.

[0016] The annular science fiction space is a circular immersive scene used for science fiction theme display or interactive experience.

[0017] The annular wall surface is a circular wall in the annular science fiction space used to carry the projected image and allow direct interaction on it.

[0018] The three-dimensional point cloud data is the spatial position data of each point on the surface of the annular wall. Each point has specific position information, which can accurately present the shape of the wall.

[0019] The projector is a projection device installed in the annular science fiction space used to project the prepared or real-time generated image onto the annular wall to create an immersive visual effect.

[0020] The spatial pose information is the position and orientation data of each projector in space, including specific position coordinates, rotation angles and optical parameters, used to determine the positional relationship between the projector and the wall.

[0021] The annular wall surface point cloud model is a digital model made according to the three-dimensional point cloud data, which can digitize the shape, concave-convex and other geometric characteristics of the annular wall.

[0022] The pose parameter data set is the parameters recorded after calibration of each projector, including position coordinates, direction angles and projection geometric relationships, which are key data that enable accurate projection of images onto the annular wall.

[0023] In the annular science fiction space, three-dimensional scanning technology is used to obtain the spatial geometric information of the annular wall. Laser scanning, structured light scanning or multi-view stereo vision reconstruction can be selected to perform complete coverage scanning of the annular wall, and finally obtain a discrete point set of the annular wall surface in three-dimensional space, i.e. three-dimensional point cloud data. These point cloud data contain the spatial coordinates of each sampling point on the wall, which can accurately reflect the shape, curvature and height distribution of the wall. At the same time, the three-dimensional coordinates and orientation angles of each projector installation position in the annular science fiction space are measured and recorded, including yaw, pitch and roll angles, forming the spatial pose information of each projector to describe the absolute position and direction of the projector in space.

[0024] Based on the collected three-dimensional point cloud data of the ring-shaped wall surface, the point cloud is filtered and registered. First, the noise points and repeated points are removed, and then the reference coordinate system is unified, and the ring-shaped wall surface point cloud model is constructed to present the overall shape and surface features of the wall in a digital form. On this basis, the pose information of each projector is mapped into this point cloud model, and geometric calibration is performed using the ray projection principle, that is, according to the optical center position and projection direction of the projector, the projection area of each projector on the wall point cloud model and the projection coordinate correspondence are calculated. By comparing the actual projection of the projector with the spatial corresponding points of the wall model, the position, direction and projection parameters of the projector are adjusted, and the projection calibration is completed. After calibration, the pose parameter data set of each projector is obtained, which includes the position coordinates, direction angles and projection geometric relationship of the projector in the wall coordinate system, which can be used to accurately map the preset picture or dynamically generated picture to the ring-shaped wall surface.

[0025] Step S102: Based on the pose parameter data set, control each projector to map the real-time state of each pixel in the preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface.

[0026] The preset picture sequence is a set of continuous projection pictures designed in advance, and the core function is to present dynamic visual effects on the ring-shaped wall surface. It includes picture frames, that is, each image or video frame in the sequence, which contains color, brightness and texture information; display order, frames are arranged in time order, and the continuous animation or interactive picture is obtained by splicing; visual effect setting, each frame can add specific animation, and element position adjustment, size change, transparency change or color change details, so that the ring-shaped wall surface can present dynamic and continuous visual effects.

[0027] Real-time state is the color, brightness or transparency information of each pixel in the picture frame at the current time point, which aims to accurately map to the actual projection position. It includes pixel color information, such as RGB or RGBA color data; and brightness or grayscale value. With this information, the projector can accurately control the display effect of each pixel on the wall, making the dynamic picture play coherent and real.

[0028] The corresponding projection coordinates are the accurate correspondence of each pixel in the picture sequence to the spatial coordinates of the ring-shaped wall surface. Both three-dimensional coordinates of the wall, such as X, Y, Z coordinates, and two-dimensional mapping coordinates, such as u, v coordinates, are included, and these coordinates have been corrected for distortion and offset to ensure accurate position.

[0029] Based on the pose parameter dataset, the spatial position and orientation information of each projector is matched with the three-dimensional coordinates of the ring-shaped wall surface to establish the mapping relationship from the local coordinate system of the projector to the three-dimensional coordinate system of the wall surface. By using the ray projection or projection matrix transformation method, the two-dimensional coordinates of each pixel in the picture frame are converted into three-dimensional projection points in the wall coordinate system, and the optical parameters of the projector such as focal length and lens distortion coefficient are integrated to correct the mapping results, ensuring the accuracy of the projection on the actual wall surface.

[0030] In the multi-projector splicing scene, the projection area of each projector on the wall surface needs to be overlapped. According to the relative position of the projectors in space, the boundary of the area projected by each projector on the wall surface is calculated, and in the overlapping area, a brightness fusion algorithm and a color balance method are used to ensure the color, brightness and geometric continuity between the spliced pictures of multiple projectors. For local curvature or slight deformation of the wall surface, coordinate correction is carried out in combination with the projection feedback image, the actual wall contour is compared with the theoretical mapping coordinates, and the projection coordinates are adjusted to eliminate the deviation, and then the final calibration projection coordinates are generated. Finally, the real-time state of each pixel is matched with the corresponding calibration projection coordinates, and the projector projects light on the ring-shaped wall surface according to the mapping, and the current frame picture is dynamically displayed. With the continuous playback of the picture sequence, each frame of pixel is updated in turn, realizing the continuous dynamic projection of the entire preset picture sequence on the ring-shaped wall surface. For example, a set of picture sequences are preset, each frame records the wave height and color change of the sea surface at that time, forming a continuous dynamic effect. Based on the pose parameter dataset of the projector, the real-time state of each pixel in each frame of picture is mapped to the corresponding projection coordinates on the ring-shaped wall surface, so that the dynamic sea surface effect with ups and downs over time and rippling light is presented on the wall surface.

[0031] Step S103: If user interaction data of the user on the ring-shaped wall surface is detected, an interaction area is determined based on the user interaction data and corresponding logic is executed.

[0032] User interaction data is behavior data collected by the interaction perception system when the user performs touch, slide, pointing or approaching actions on the ring-shaped wall surface. It includes touch data such as touch position coordinates, touch area, touch pressure and touch duration; and slide data such as slide start position, slide path, slide direction and slide speed.

[0033] The interaction area is the wall surface area where the user is currently operating, determined by combining the position of the user interaction data on the ring-shaped wall surface, and comparing the spatial coordinate system of the ring-shaped wall surface and the projection mapping relationship.

[0034] The corresponding logic is to identify the interaction area and implement the corresponding response strategy or content update process to achieve the interaction effect. It can include content interaction logic responsible for starting content switching, object movement, element selection, and other functions; dynamic effect logic can trigger area-related water wave diffusion, particle response, light effect following, and other special effects; projection control logic updates the projection coordinate mapping of the interaction area to synchronize the response content and user operation position accurately.

[0035] When running in the annular science fiction space, a multi-source perception pipeline will be started to monitor user interaction behavior on the annular wall. Depth cameras, structured light cameras, and thermal imaging auxiliary sensors installed on the top of the space or the edge of the wall will collect user motion data covering the entire annular wall in real time, forming a sequence of original user interaction data frames containing hand trajectory, body posture, fingertip contour change, and touch depth change. Use background modeling algorithms such as Gaussian Mixture Model to eliminate static backgrounds, and then use human key point estimation algorithms such as OpenPose or MediaPipePose to identify user upper limb key point positions, and calculate the shortest distance from each key point to the wall point cloud model based on depth information to determine whether the user is touching or approaching the wall projection area. If the distance between the hand key point and the wall model is below the set threshold, an interaction event trajectory will be marked in the current frame, and then the direction, speed, and acceleration information of the user's hand will be extracted using optical flow estimation methods such as PWC-Net or FarnebackOpticalFlow to organize structured user interaction data.

[0036] After obtaining the user interaction data, the three-dimensional position of the hand in the space coordinate system will be projected onto the point cloud model of the annular wall. Specifically, the intersection of the user's hand contact point and the point cloud surface is calculated using the normal vector field of the wall point cloud, and then this intersection is converted to the annular wall unfolding coordinate system, such as angle θ and vertical height z in cylindrical coordinates. Based on the position, trajectory direction, motion speed, and motion type in this coordinate system, such as tapping, sliding, pressing, waving, and hitting, a rule-based motion classifier or a Transformer-based motion recognition model is used to infer the interaction semantics, and further determine whether the point is in a pre-defined interaction area. The interaction area is composed of multiple wall geometry segments or texture blocks, and each area has a unique spatial index range, for example, a certain area may correspond to the interval of wall angle 45° to 60° and height 1.0 meters to 1.6 meters. Comparing the projected coordinates of the user interaction point with these region index ranges can determine the current triggered interaction area.

[0037] After locking the interactive region, the system will execute the pre-set corresponding logic according to the type of the interactive region and the current action semantics. For example, when the region type is a sea surface dynamic wave region and the action semantics is a fast waving, the system will call the physical wave simulation system in the real-time graphics rendering pipeline, such as a GPU computing model based on the shallow water equation, input the velocity vector of the user waving as a disturbance matrix of the wave field, and update the wave height map in real time. Then, using the extrinsic parameters of each projector in the pose parameter data set, the updated wave texture is mapped back to the projection coordinate system of each projector to refresh the projection picture of the corresponding region, so that the water surface splashing or wave diffusion effect synchronized with the user's action appears on the ring-shaped wall surface.

[0038] On the basis of the above technical solutions, optionally, if the user interaction data of the user on the ring-shaped wall surface is detected, the interactive region is determined based on the user interaction data and the corresponding logic is executed, including: If the user interaction data of the user on the ring-shaped wall surface is detected, the touch position data of the user is continuously extracted based on the corresponding relationship between the user interaction data and the current projection picture of the ring-shaped wall surface in continuous time; If the touch positions determined based on the touch position data in continuous time all correspond to the water body interactive region of the current projection picture, the touch force data of the user is continuously extracted based on the user interaction data; Based on the touch position data, the local neighborhood point set is continuously determined in the ring-shaped wall surface point cloud model with the touch position as the center, and the local surface fitting is performed on the local neighborhood point set to obtain the local fitted surface corresponding to the touch position; The local surface characteristics and neighborhood topological structure of the touch position are continuously extracted based on the local fitted surface, and the local coordinate system of the touch position is established based on the local surface characteristics and neighborhood topological structure, and the local geometric information of the touch position is continuously determined based on the local coordinate system; The SPH simulation parameters are dynamically configured based on the local geometric information and the touch force data, the current state data of the water body interactive system is continuously obtained, and the SPH simulation parameters are continuously updated based on the current state data; Based on the continuously updated SPH simulation parameters, the acceleration and motion trajectory of each floating interactive object in the local gravitational field of the water body interactive system are calculated, and the dynamic state data and real-time position data of each floating interactive object are determined based on the acceleration and motion trajectory; Based on the dynamic state data and real-time position data, the dynamic projection coordinates of each floating interactive object on the ring-shaped wall surface are calculated, and each projector is controlled to project and display on the ring-shaped wall surface based on the dynamic projection coordinates until the user interaction data of the user on the ring-shaped wall surface cannot be detected.

[0039] In this scheme, the current projection picture is the visual state of the picture presented by the projector on the ring-shaped wall at the moment, including the brightness, color, and texture information of each pixel, as well as the rendered interactive effects corresponding to the user's actions.

[0040] The corresponding relationship is the mapping relationship between the user's interactive actions in time and space and the pixels or regions of the current projection picture, which can clearly determine the user's touch, sliding, and other operations and the specific position of the picture on which the operations are performed.

[0041] Touch position data is a set of three-dimensional spatial coordinates of the contact point detected when the user interacts on the ring-shaped wall, including the change information in continuous time.

[0042] Touch position is the point or region that the user specifically touches or approaches on the wall, which is a single element in the touch position data.

[0043] Water body interactive region is a region in the projection picture with liquid simulation or ripple response function. As long as the user touches this region, the water body interactive effect will be triggered, such as ripple diffusion, floating interactive object splashing, and other performances.

[0044] Touch force data is the numerical information of the force generated when the user touches, slides, or presses the wall, including the size, direction, and action time of the force. These data are used to calculate liquid dynamics or simulate ripple disturbance effects.

[0045] Local neighborhood point set is a set of surrounding three-dimensional points selected from the point cloud model of the ring-shaped wall centered on the touch position, used for local surface fitting and topology analysis.

[0046] Local fitting surface is a continuous surface generated by a surface fitting algorithm on the local neighborhood point set, such as least squares, Bezier fitting, or spline fitting algorithms. The purpose is to approximately restore the geometry of the wall surface around the touch position.

[0047] Local surface characteristics are the geometric properties of each point on the local fitting surface, including normal vector, curvature, and inclination, which are used to determine the local surface morphology and the direction in which the interactive effect should be presented.

[0048] Neighborhood topology structure is the spatial connection relationship and distribution pattern between the local neighborhood point set, such as the distance between points, local grid topology, and the relative order of points.

[0049] Local coordinate system is a three-dimensional coordinate reference system established based on the touch position and its surrounding local neighborhood geometry, used to describe the precise position and motion direction of the touch point and floating interactive object in the local space.

[0050] Local geometry information is the three-dimensional space information of touch position in local coordinate system, including specific position, local surface normal, curvature, and relative position of neighborhood, which is used to calculate the boundary conditions and local mechanical response of liquid simulation.

[0051] SPH simulation parameters are a set of parameters for fluid calculation of smooth floating interactive objects, including floating interactive object mass, initial velocity, viscosity coefficient, fluid density, external force, etc., which are the key to driving water interaction simulation.

[0052] Water interaction system is a real-time calculation system that simulates the response of wall water, based on SPH or other fluid dynamics models, which can generate effects such as ripples, ripples, floating interactive object splashing, etc., and also output floating interactive object state and dynamic rendering data.

[0053] Current state data is the real-time physical state information of each floating interactive object in the water interaction system, including position, velocity, acceleration, and force, etc., which is used to update the simulation calculation and projection display content.

[0054] Floating interactive object is a virtual or physical interactive element that can produce displacement, rotation, up and down floating response in the water interaction area with local water surface disturbance.

[0055] Local gravitational field is the local force field acting on the water floating interactive object, including gravity, user applied touch force, and simulated liquid internal force.

[0056] Acceleration is the instantaneous velocity change rate vector of the floating interactive object under the action of the local gravitational field, which can describe the speed and direction of the movement change of the floating interactive object after being forced.

[0057] Motion trajectory is the three-dimensional space path formed by the floating interactive object over time, representing the moving trajectory of the floating interactive object in the wall liquid simulation.

[0058] Dynamic state data is the comprehensive information of the floating interactive object in the water interaction system over time, including acceleration, velocity, position, and force, which is used to render dynamic effects of liquid.

[0059] Real-time position data is the three-dimensional space coordinates of the floating interactive object at the current time point, through which the water simulation effect can be mapped to the projection coordinates of the ring-shaped wall.

[0060] Dynamic projection coordinates are the final projection position coordinates of the floating interactive object or interactive element on the ring-shaped wall, which are directly used to control the projection of the projector after combining with the calibrated projection coordinates, to realize real-time interactive effect.

[0061] When the user's interaction data on the ring-shaped wall surface is captured, the current projection picture is first obtained, which is formed by the cooperation of multiple projectors. Under the support of three-dimensional point cloud model and projection geometric calibration, the three-dimensional landing point position, texture content and corresponding time frame of each pixel on the wall surface in the picture are accurately recorded. To determine the specific picture area touched by the user, the touch trajectory captured by the depth camera is spatially overlapped with the three-dimensional landing points of the projection picture, and the corresponding relationship between the touch trajectory and the projection content is established. This relationship is determined through nearest neighbor three-dimensional point matching and time consistency filtering, ensuring that the real position touched by the user can be accurately determined. In a continuous time period, the touch position data is continuously extracted according to the touch trajectory detected by the depth camera, and after jitter compensation, stability judgment and surface fitting analysis, the real wall contact point, i.e. the touch position, is finally obtained.

[0062] When it is confirmed that the projection pixel corresponding to the touch position belongs to the water body interaction area, subsequent processing can be performed; this area is a preset interactive liquid area in the projection, and its range is defined by judging whether the touch landing point is located in the three-dimensional region mask volume. Since the system will first collect user interaction data of the touch area when detecting the interaction between the user and the ring-shaped wall surface, this data is composed of a sequence of depth frames, a sequence of RGB images, and the spatial position record of the touch area in a continuous time, which collectively reflects the contact behavior of the user's hand or object with the projected water body area. After obtaining the data, the touch force characteristics are analyzed: the depression amplitude of the touch point on the time axis is calculated using consecutive depth frames, and an increasing depth value means that the touch point applies a greater vertical force to the wall surface; at the same time, the contour of the touch area in the depth map is tracked, and by calculating the expansion or contraction of its area over time, different pressure modes such as fingertip point pressure, palm coverage, and rapid sliding pressure are distinguished. Optical flow is solved for the image sequence, and by detecting the continuous displacement of the texture around the touch point, the boundary deformation direction and the brightness distribution flow trend, the tangential pushing direction and sliding force of the user's force are inferred. By continuously aligning and fusing the depth pressure characteristics, contact area change characteristics and optical flow direction characteristics, touch force data is generated.

[0063] After detecting the touch position, a spatial neighborhood search is performed in the three-dimensional point cloud model of the ring-shaped wall surface, with the three-dimensional spatial coordinates of the position as the center. The point cloud data is preprocessed using a k-d tree or octree index structure for fast querying. During the query process, the point cloud distribution around the touch center point is searched with a set spatial radius or a specified number of neighboring points as the condition, and the points that meet the distance requirements are collected as a local neighborhood point set. After obtaining the local neighborhood point set, the points are used to construct a local surface. First, the covariance matrix of the point set is calculated, and the local least squares plane is obtained through eigenvalue decomposition. Then, the plane is used as the initial estimate, and the least squares surface fitting, moving least squares smoothing surface technique, or cubic B-spline surface fitting method is used to generate a continuous and differentiable local fitting surface on the point set distribution. After the fitting is completed, multiple sampling points are selected on the surface, and the normal direction, principal curvature, secondary curvature, curvature gradient distribution of each sampling point are calculated, and the direction consistency of the curvature principal direction on the surface is further identified. These calculation results together constitute the local surface characteristics of the touch area. After completing the surface characteristic calculation, the neighborhood topology structure is extracted from the original neighborhood point set. Specifically, the local neighborhood point set of each point is determined through K-neighbor search or radius neighborhood search technology, and the local neighborhood graph structure is constructed according to the spatial connectivity of the point cloud and the preset distance weight. Based on this graph structure, the local density of each point, the connection relationship with adjacent points, the local connectivity, and the gradient trend of the distance change between points can be calculated. These calculations involve kernel density estimation to evaluate local density, graph Laplacian to analyze local connectivity, and local linear fitting or principal component analysis to determine the gradient trend of distance change.

[0064] When the local surface characteristics and neighborhood topology structure are completed, a local coordinate system of the touch area is established based on them, with the normal vector of the fitted surface at the touch position as the Z-axis, the principal direction corresponding to the principal curvature as the X-axis, and the Y-axis direction determined by the cross product of the X-axis and the Z-axis, so that the three directions form a right-handed orthogonal coordinate system. This coordinate system can stably present the local surface orientation and structural changes of the touch area. Subsequently, the original position of the touch point in the global point cloud coordinate system, the normal direction on the local fitted surface, the curvature value, the principal direction vector of the curvature, the neighborhood density and connectivity, and other information are all projected and converted into this local coordinate system. After conversion, all geometric properties are re-expressed as geometric quantities relative to the local surface. These converted data, which are the local geometric information, contain the real surface shape, local bending degree, direction change, neighborhood distribution characteristics, and other boundary constraint parameters that can be used for water body simulation.

[0065] After obtaining the local geometric information and touch force data, the two types of information are used to dynamically configure the SPH simulation parameters in the water body simulation. The normal direction, curvature value, surface concave-convex feature, neighborhood density, and local coordinate system direction in the local geometric information are input into the SPH boundary processing flow, so that the boundary stress, normal reflection intensity, and disturbance decay speed of the water surface can be adjusted according to the real surface morphology. The vertical penetration force size, tangential sliding force, action direction, and duration in the touch force data are used to control the local pressure source intensity, flow velocity gain, and disturbance propagation radius of the particle neighborhood, so that the SPH simulation can accurately reflect the user force application mode. In the simulation initialization stage, the two types of data are also used to adjust the key SPH simulation parameters such as the kernel function support radius, local pressure term coefficient, viscosity term weight, and fluid surface tension, to ensure that the water body response is consistent with the touch action.

[0066] After the dynamic configuration of the SPH simulation parameters is completed, the current state data in the water body interaction is read in each simulation cycle. These data are directly generated from the simulation results of the previous frame, including the position, velocity, orientation, fluid force received, flow field velocity distribution, and internal pressure gradient of all floating interactive objects on the water surface. By traversing the simulation grid and particle storage structure, these numerical values are extracted one by one to drive the next round of SPH update. According to the pressure gradient change, viscosity change, boundary condition change, and disturbance state caused by user touch in the flow field, the pressure term parameters, velocity correction parameters, and surface energy compensation parameters are adjusted again, so that the SPH simulation parameters are continuously updated on the time axis, realizing the stable evolution of the dynamic water body. After obtaining the latest SPH simulation parameters, the force status of each floating interactive object in the local gravitational field is calculated in the current simulation step: the immersed volume, contact area, buoyancy direction and size, fluid dynamics from the SPH flow field, including pressure force, viscous damping force, tangential flow traction force, and disturbance force caused by user force application of the object are comprehensively solved, and then based on the Newton's law of motion, the total force is divided by the object mass to obtain the corresponding instantaneous acceleration; then in the time integration process, the semi-implicit Euler integration or Runge-Kutta integration method is used to update the motion trajectory of the object with time, including the translation trajectory, up-down floating trajectory, and rotation trajectory around the center of gravity.

[0067] After the acceleration and trajectory calculation is completed, the dynamic state data and real-time position data of each floating interactive object are obtained, the former includes the instantaneous speed, angular speed, attitude change, force direction and force size of the object, and the latter is the updated three-dimensional position coordinates in the world coordinate system. In order to correctly display the floating interactive object on the ring wall, the real-time position data is input into the point cloud geometry of the wall, and the three-dimensional position of the object is converted into the two-dimensional imaging plane coordinates of the corresponding projector through the geometric transformation from three-dimensional space to projection plane, to generate dynamic projection coordinates. This conversion process combines the calibrated pose of the projector, the curvature of the wall, the spatial depth of the object and its relative mapping relationship in the local coordinate system, to ensure that the projection result is accurately aligned with the real position. After the dynamic projection coordinates are generated, they are directly transmitted to each projector, so that the projector updates the display content in real time according to these coordinates, and the motion state of the floating interactive object is accurately projected onto the corresponding area of the ring wall. This simulation, update and projection process is continuously executed until no new user interaction data is detected, at which time the SPH boundary disturbance gradually decays, and the floating interactive object slowly returns to a calm state as the water body naturally evolves. For example, when the user drags a boat in the water area of the ring wall, the boat will move, rotate and float up and down with the gesture, while the surrounding water surface will form a ripple diffusion, and the projection picture will display the position change of the boat and the water disturbance effect in real time.

[0068] On the basis of the above technical solutions, optionally, if user interaction data of a user on the ring wall is detected, an interactive region is determined based on the user interaction data and corresponding logic is executed, further comprising: If user interaction data of a user on the ring wall is detected, touch position data of the user is continuously extracted based on the correspondence between the user interaction data and the current projection picture of the ring wall within a continuous time; If the touch positions determined based on the touch position data within the continuous time all correspond to the starry sky interactive region of the current projection picture, touch force data and sliding direction data are continuously extracted based on the user interaction data; The touch force data is mapped into a rotation field radius and a first rotation angular velocity in real time through a nonlinear function, and a local rotation field is generated in the starry sky interactive region with the touch position as the center based on the rotation field radius and the first rotation angular velocity; Based on the touch position data and the rotation field radius, lines in a preset initial line set of the starry sky interactive region that are located within the range of the local rotation field are screened to obtain a target line set that enters the local rotation field; Based on the local rotation field and the touch sliding direction, the rotation direction and the second rotation angular velocity of each line in the target line set are determined; The first dynamic motion state of each line is obtained by applying a rotation effect on each line based on the rotation direction and the second rotation angular velocity and superimposing a bending motion driven by the original Perlin noise; The display color of each line is determined based on the first dynamic motion state and a preset color mapping rule; The real-time rendering data of each line is generated based on the first dynamic motion state and the display color, and the projectors are controlled to project and display on the ring-shaped wall surface based on the real-time rendering data until no user interaction data of the user on the ring-shaped wall surface can be detected.

[0069] In this scheme, the starry sky interaction region is a virtual starry sky region predefined on the ring-shaped wall surface and responsive to user touch, containing starlight lines or nebula patterns, and the user touch of the region can trigger corresponding visual interaction effects.

[0070] The sliding direction data is direction information generated by the user sliding on the touch screen or the ring-shaped wall surface, used to indicate the moving direction of the user operation, covering horizontal, vertical or oblique displacement vectors.

[0071] The nonlinear function is a mathematical mapping function used to convert the touch force data applied by the user into rotation field parameters such as radius and angular velocity. Its variation and input are not in a linear relationship, which can enhance the natural feeling and controllability of the interaction.

[0072] The rotation field radius is the effective influence range radius formed around the touch point in the starry sky interaction region, acting on the surrounding starry sky lines, and determining which lines will be affected by the rotation field.

[0073] The first rotation angular velocity is the angular velocity of the initial action of the rotation field, calculated from the touch force of the user, used to control the rotation speed of the lines around the touch point.

[0074] The local rotation field is a rotation influence region formed around the user touch position, acting on the starry sky lines within the local range, allowing these lines to rotate around the center point and present dynamic visual effects.

[0075] The preset initial line set is an initial set of all starlight lines or trajectories to be rendered in the starry sky interaction region of the ring-shaped wall surface, containing the starting position, length and initial motion state of each line.

[0076] The target line set is a subset of lines selected to participate in the rotation effect within the action range of the local rotation field, and only these lines will respond to the influence of the rotation field.

[0077] Rotation direction is the direction of the target line rotating around the touch point, usually calculated according to the user's sliding direction and the rotation force field, divided into clockwise or counterclockwise.

[0078] Second rotation angular velocity is the angular velocity of the target line actually rotating in the local rotation force field, which will be adjusted based on the first rotation angular velocity and the sliding direction, so that the line movement is more consistent with the user's operation.

[0079] The first dynamic motion state is the real-time dynamic performance of the target line under the action of the rotation force field, including the rotation angle, the bending shape, and the bending motion state superimposed with the original Perlin noise.

[0080] Real-time motion speed is the instantaneous speed of each line in the rotation and bending motion of the current frame, which is used to calculate the color change and rendering effect.

[0081] The preset color mapping rule is a color mapping scheme preset according to the motion speed, rotation angle or other dynamic characteristics of the line, which is used to convert the dynamic motion state into the corresponding visual color effect.

[0082] The display color is the final color value of the line in the current rendering frame, which is determined by the real-time motion speed and the color mapping rule, and is used for visual presentation.

[0083] Real-time rendering data contains the spatial position, rotation angle, bending shape and display color of each line, which is used to drive the projector to perform real-time projection display on the ring-shaped wall surface.

[0084] After confirming that the projection pixel corresponding to the touch position belongs to the starry sky interactive area, first analyze the touch point's depth change by depth frame to judge the vertical force size; At the same time, analyze the outline change of the touch area to see how the contact area is expanded, so as to distinguish whether the touch is a fingertip point pressure, a palm covering or a sliding contact. Then use the optical flow tracking method to analyze the moving direction and speed of the local texture in the continuous frame to get the sliding direction of the touch. Fuse the features such as depth indentation, contact area change and optical flow direction in time sequence to generate touch force data and sliding direction data, which respectively indicate the size, direction, duration and moving direction of the force. Use the manually set nonlinear mapping function to convert the touch force data into the parameters of the rotation force field. The function will calculate the action radius of the rotation force field, which is the rotation force field radius, according to the force size, duration and finger sliding speed, and will also map the force size into the initial rotation rate, which is the first rotation angular velocity. The mapping function can be Sigmoid function or power law function, so that small touch force corresponds to small rotation radius and angular velocity, and large force can produce stronger rotation effect, ensuring that the interactive experience is coherent and natural.

[0085] With the user touch point as the center, first do a spatial search in the three-dimensional point cloud model of the ring wall surface, calculate the Euclidean distance of each point to the touch point. All points with a distance less than the radius of the rotating force field are marked as affected areas, forming a local rotating force field. In order to determine which starry sky lines will be affected by the force field, the vertex coordinates of each line in the preset initial line set are traversed, and the Euclidean distance between the three-dimensional position of the vertex and the center point of the force field is compared. Only the line whose vertex at least falls in the affected area will be selected into the target line set. After determining the target line, the rotation direction of each line is calculated by vector operation. Specifically, the vector from the center of the force field to the vertex of the line is calculated, and the cross product or angle included angle between the vector and the sliding direction vector of the user's finger is calculated to obtain the rotation direction. The second rotation angular velocity is obtained by weighting and combining the distance of the line to the touch point, the rotating force field intensity and the finger sliding speed. The closer the distance to the center and the faster the sliding speed, the greater the rotation rate. After determining the rotation direction and the second rotation angular velocity of each target line, each vertex on the line is dynamically adjusted. First, analyze the position of each vertex relative to the center of the local rotating force field, including the distance to the touch point and the spatial direction, and then plan the rotation path of the vertex along the rotation direction. The vertex will slowly rotate along the path at the second rotation angular velocity. The vertex close to the touch point rotates a large angle, and the vertex far away rotates a small angle, which can ensure that the line as a whole presents a natural rotation arc.

[0086] In the rotating process, in order to make the line have soft bending and random swing effect, Perlin noise is used to generate a small random offset in three-dimensional space for each vertex. Specifically, the current position of the vertex is input into the Perlin noise function to obtain a random disturbance in the form of a three-dimensional vector, and then the disturbance intensity is adjusted according to the distance of the vertex to the touch point, so that the part close to the touch point has large bending amplitude and the part far away has small bending amplitude. The rotated vertex position and noise disturbance vector are superimposed to make the vertex complete rotation and have a small swing and bending, and present a real and natural dynamic effect. After repeating the above operation on all vertices of the line, the line as a whole will present a comprehensive state of rotation, bending and swinging, which is the first dynamic motion state of the line. The three-dimensional position, rotation direction and bending amplitude of each vertex are updated on the time axis, and when the user touches the starry sky interaction area, the line can rotate, swing and twist with the gesture, presenting a smooth and natural dynamic motion effect, which provides basic data for subsequent real-time speed calculation, color mapping and projection display. After obtaining the first dynamic motion state of each line, the three-dimensional position and rotation angle of each vertex on the line in the continuous time frame are tracked. Specifically, the position change of each vertex between the current frame and the last frame is compared, including the spatial translation displacement and the rotation angle change, and the instantaneous velocity vector of the vertex is calculated through these changes, and then the velocity vectors of all vertices are integrated to obtain the real-time motion speed of each line as a whole. The speed of the line not only reflects the rotation rate, but also contains the displacement component caused by bending and swinging, which completely describes the dynamic change of the line.

[0087] After obtaining the real-time motion speed, the speed information of the line is converted into visual color according to a preset color mapping rule. For example, the line with low motion speed is mapped into cold color tone such as blue or purple, and the line with high motion speed is mapped into warm color tone such as orange or red. Meanwhile, the rotation angle and the bending amplitude are combined to fine-tune the color, so that the color presents a gradient effect, and the dynamic strength and motion state of the line are intuitively reflected. The color information is recorded as the display color attribute of each line. The first dynamic motion state of the line and the corresponding display color are integrated to generate real-time rendering data. The rendering data includes the three-dimensional coordinates of all vertices of each line, the rotation angle, the bending amplitude and the color information, forming a complete set of geometric and visual attributes that can be used for display. In the real-time rendering stage, the projection control reads the rendering data, maps the three-dimensional position and dynamic effect of the line to the projection coordinates of the ring-shaped wall surface, and adjusts the visual color of each line according to the display color. The projection is continuously refreshed at a high frequency, so that the starry line on the ring-shaped wall surface follows the sliding, rotating and touching actions of the user's finger, presents smooth and dynamic rotation, bending, swinging and color change effects, and forms an intuitive interactive visual experience. If the touching or sliding behavior of the user on the ring-shaped wall surface cannot be detected, the rendering data stops updating, and the projection effect remains unchanged or slowly returns to the initial state, ending the interactive display. For example, when the user slides the finger on the ring-shaped wall surface to touch the starry area, the starry line around the touching point can be seen to slowly rotate in the direction of the finger sliding, the line close to the finger rotates faster, and the line far away rotates slightly slower, and meanwhile, the line slightly bends and swings, presenting a natural undulating feeling. As the finger moves, the line being rotated and bent changes position along the trajectory of the finger, spreads outward, and the color of the line also changes with the rotation speed, gradually transitioning from cold color to warm color, so that the entire starry sky presents a dynamic visual effect of flowing, rotating and flickering.

[0088] On the basis of the above technical solutions, optionally, if the user interaction data of the user on the ring-shaped wall surface is detected, the interactive region is determined based on the user interaction data, and the corresponding logic is executed, further comprising: If the user interaction data of the user on the ring-shaped wall surface is detected, the touching position data of the user is continuously extracted based on the correspondence between the user interaction data and the current projection picture of the ring-shaped wall surface in continuous time; If the touching positions determined based on the touching position data in continuous time all correspond to the text interactive region of the current projection picture, the text node with a distance from the touching position satisfying a preset distance threshold is selected from a preset initial text set of the text interactive region as a target text node triggered currently based on the touching position data; Based on the local spring network corresponding to the target text node, an associated text node set with a connection strength exceeding a preset elasticity threshold is selected from the target text node; Obtain the node position and elastic link direction of each associated text node in the associated text node set, and determine the force direction and force magnitude of each associated text node based on the touch action based on the node position and elastic link direction. Based on the force direction and magnitude, an elastic displacement perturbation is applied to each associated text node in the associated text node set to obtain the second dynamic motion state of each associated text node. Based on the second dynamic motion state, the positions of each associated text node are updated to obtain the text node update result after elastic response, and each projector is controlled to perform real-time rendering and display based on the text node update result until the user interaction data on the circular wall can no longer be detected.

[0089] In this solution, the text interaction area is a touchable area on the circular wall that displays text content. When a user touches this area, the dynamic interactive effects of the text nodes are triggered.

[0090] The preset initial text set is a collection of all text nodes that are pre-arranged in the text interaction area, including the position, font, and number information of each text.

[0091] The preset distance threshold is the maximum range of touch influence set to determine the spatial distance between the user's touch point and the text node. Only text nodes within the threshold will be triggered for interaction.

[0092] A text node is a single interactive text unit within a text interaction area. It has its own position coordinates, possesses the property of elastic movement, and can also respond to user touch operations.

[0093] The target text node is the text node that is within a preset distance threshold from the touch position when the user touches it; that is, the text node that is currently being triggered and interacted with.

[0094] Local spring networks are text node interconnection structures built around target text nodes. Each connection has elastic properties to simulate the elastic interaction between text nodes.

[0095] Connection strength refers to the degree of elasticity between text nodes. The higher the value, the more obvious the elastic effect between the two nodes.

[0096] The preset elasticity threshold is the minimum strength standard for filtering elastic links. Only text nodes with a connection strength exceeding this threshold will participate in the force response calculation.

[0097] The associated text node set is a group of text nodes connected to the target text node by elastic links. Only those text nodes with a connection strength exceeding the preset elastic threshold are included in this set. These nodes will be affected by forces during the interaction process.

[0098] The associated text node is a single text node in the associated text node set. When the user touches it, it will produce elastic displacement.

[0099] The node position is the three-dimensional coordinate of the text node on the ring wall, mainly used to calculate the force direction, and will also be used as the basis for updating the display position.

[0100] The elastic link direction is the direction vector of the elastic connection between text nodes, which determines the specific direction of force transmission.

[0101] The force direction is the direction of the force generated by the user's touch, transmitted to the associated text node through elastic links. This direction directly determines the movement direction of the text node.

[0102] The force size is the strength of the elastic force acting on the text node, related to the connection strength between nodes and the force of the user's touch. It is mainly used to calculate the amplitude of displacement.

[0103] The second dynamic motion state is the real-time motion of the associated text node after being subjected to force, including position offset, velocity direction, and dynamic elastic response. These information will be used to render the corresponding animation effect.

[0104] The text node update result is the new position and state information of the text node after elastic response and position update. These information are used to control the real-time display of the projector on the ring wall.

[0105] After confirming that the projection pixel corresponding to the touch position belongs to the text interaction area, the target text node is selected from the preset initial text set. This initial text set contains the three-dimensional position and attribute information of each text node in the text interaction area. Only the nodes with a spatial distance within the preset threshold are left, and they are marked as the target text node triggered by the touch point.

[0106] After determining one or more target text nodes, the connection relationship of these nodes in the local spring network is analyzed. Each target text node and the surrounding text nodes are connected by elasticity, and each connection has a numerical strength attribute. Traverse all the elastic connections of the target text node, read the strength value of each connection, and compare it with the pre-set elastic threshold. Only when the connection strength exceeds the threshold, the corresponding connected text node will be selected into the associated text node set, so as to ensure that only the nodes obviously affected by elasticity participate in the subsequent dynamic response. This operation will be executed in real time at each time frame to adapt to the continuous change of the user's touch position. The establishment of the local spring network starts from the spatial position and association relationship of the text nodes, and simulates the elastic connection between the texts. Specifically, first, the initial position coordinates of each text node in the text interaction area in the three-dimensional space are calibrated to form a node set. Then for each node, the connection edges between the text nodes adjacent or semantically related to it are established according to the spatial distance or logical relationship, each connection edge is simulated as a spring, and an initial elasticity coefficient or connection strength is assigned. This coefficient can be determined according to the distance between nodes, the degree of semantic association or user-defined design parameters. The closer the distance between nodes, the stronger the semantic association, the greater the connection strength, and the spring is more likely to transmit force. After establishing the connection, each text node will form an elastic link with several surrounding nodes, and these links can simulate the stretching and compression of the spring when the node is stressed. At the same time, a direction vector is defined for each elastic connection, pointing to the position of the connected node, and the direction of force transmission is clear. In this way, each node not only knows which nodes it is connected to, but also knows the path of force action when calculating stress. Finally, all nodes and their elastic connections together constitute a local spring network, and when the user touches a node, the force will be transmitted to the adjacent nodes along the network, forming a chain of dynamic response effects.

[0107] After the target text node and its local spring network are determined, all elastic connections of the target text node are traversed first, and the connection edges with connection strength exceeding the preset elastic threshold are screened out. The connected text node set corresponding to these connection edges is defined as the associated text node set. To obtain the position of each associated text node, the coordinate records of the associated text nodes in the three-dimensional space are read. These coordinates are the actual projection positions of the text on the ring-shaped wall surface or the positions calibrated in the three-dimensional modeling space. The position data of each node includes the spatial coordinates of X, Y and Z directions, which are used to describe the accurate position of the node. According to the definition of the elastic connection, the direction vector of each edge is obtained. This vector points from the target text node to the associated text node and is used to represent the direction of elastic force transmission. Specifically, the spatial coordinates of the associated text node are subtracted from the coordinates of the target text node to obtain the original vector, and then the original vector is normalized to form a standardized direction vector. In this way, each associated text node not only has a clear position coordinate, but also determines the connection direction with the target text node. Next, the force direction and size of the associated text node based on the touch action are determined. The touch force data of the target text node is obtained first, including the action direction and the applied force size. Then, the action force is transmitted to the associated text node along the elastic connection direction. The force direction generally points to the associated node along the direction vector, and is corrected in combination with the angle between the action direction of the target node and the elastic connection direction, so that the force transmission conforms to the physical elastic characteristics. The force size is calculated according to the principle of elastic mechanics. The initial force applied by the target node, the elastic connection strength and the distance between the nodes are weighted. The higher the connection strength and the closer the distance between the nodes, the greater the force transmitted to the associated text node, and vice versa. This calculation is performed for each elastic edge, and the direction and size of the force borne by each associated text node in the current time frame are obtained. These data will be the basis for the elastic displacement of the associated text node. Each associated text node will form a complete set of data, including the three-dimensional space position coordinates, the force direction along the elastic connection direction and the force size. With the aid of these data, the next step can apply corresponding displacement disturbance to the associated text node, realize the dynamic elastic response of the text node, and further form a chain animation effect.

[0108] After obtaining the force direction and size of each associated text node, the elastic displacement of each node in three-dimensional space is calculated. First, the force direction vector of each node is normalized, so that its length becomes 1, which can uniformly describe the direction, and the force size will be used as the basis for displacement amplitude. Then, combined with the distance between the node and the target text node, the displacement amplitude is adjusted according to the principles of elastic mechanics. The closer the node is to the target node, the larger the displacement amplitude is scaled; the farther the node is, the smaller the displacement amplitude is scaled, to simulate the characteristic of spring force weakening with distance. The displacement increment is calculated in three-dimensional coordinates. For each node, the original position coordinates are added to the force direction vector multiplied by the adjusted displacement amplitude, to obtain the new coordinates of the node. This step completes the preliminary elastic response of the node. In order to make the animation more natural and flexible, a small random disturbance is added: a smooth changing random vector in three-dimensional space is generated using Perlin noise, the vector amplitude is not large, and it is superimposed on the node coordinates with the main displacement to make the node present slight jitter or bending effect while moving elastically, simulating the vibration and flexible swing of real objects after being forced.

[0109] After applying the main displacement and disturbance, the new three-dimensional coordinates of each node at the current time frame are recorded, and the displacement direction, amplitude, and force source information of the node are saved. These data together constitute the second dynamic motion state of the node. This state not only includes the real-time position of the node in three-dimensional space, but also covers the motion direction, amplitude, and elastic response characteristics of the node after being forced, which can be used for subsequent motion speed calculation, collision detection, or color mapping visualization processing. The whole process is repeated for each associated text node, so that the force generated by the touch on the target node is transmitted to the surrounding nodes through the spring network, forming a chain of dynamic elastic effect. Finally, the second dynamic motion state of all nodes is integrated into a real-time updated data set, which is directly used to refresh the projection display, realizing the elastic deformation animation of the text node. The second dynamic motion state of each associated text node is written into the global node position record, and the three-dimensional coordinates are refreshed. The update results of all nodes are integrated into the text node update set, which contains the new position, elastic displacement, and dynamic offset information of each node. When rendering each frame of picture, this set is read to map the three-dimensional coordinates to the annular wall projection coordinates, and the real-time rendering picture is generated accordingly. Finally, the user will see that the target text node moves along with the touch, and the surrounding associated nodes produce a chain of elastic displacement animation. This process is continuously updated until the user stops interacting or leaves the text area, at which point the touch data collection and elastic response stop, and the text node returns to the static display state. For example, when the user slides the "star" character, the "star" moves in the direction of the finger, and the connected "moon", "day", "light" and other characters are slightly displaced and swung, presenting a chain of elastic motion; after the finger stops, the "star", "moon", "day", and "light" gradually return to their original positions.

[0110] Step S104: Continuously acquire the projection feedback image of the ring-shaped wall surface in the execution of the corresponding logic process, perform dynamic calibration and micro-compensation processing based on the projection feedback image and the pose parameter data set, and generate the calibration projection coordinates along the ring-shaped wall surface and the corresponding real-time interactive response data.

[0111] The projection feedback image is a real-time image of the ring-shaped wall surface in the ring-shaped science fiction space, captured by a calibration camera, a depth camera, or a fisheye wide-angle camera.

[0112] The calibration projection coordinates are the accurate mapping coordinates of each projection pixel on the ring-shaped wall surface point cloud model, obtained by analyzing the projection feedback image and combining the pose parameters of the projector.

[0113] The real-time interactive response data is data generated by remapping based on the user's current action and the original projection content according to the calibration projection coordinates, used to drive the dynamic projection effect.

[0114] During the continuous execution of the user interaction logic, to ensure the accuracy of the ring-shaped wall surface projection, the projection feedback image of the wall surface is continuously collected. These images are captured by high-definition or depth cameras around the wall at a fixed frame rate, recording the real-time state of the wall projection, such as the actual landing point of the pixels, the brightness distribution, the deformation caused by the curvature, the misalignment of the splicing, and the changes in the shadow and outline left by the user's obstruction or touch. At the same time, the projection offset caused by the small vibration of the projector, temperature changes, or other environmental factors is also reflected in the feedback image.

[0115] The feedback image is pre-processed, including denoising, edge detection and feature point extraction, and then the observed feature points are matched with the feature points of the corresponding frames of the preset picture sequence. Then, the RANSAC algorithm is used to remove abnormal matching points, the pixel offset is calculated through perspective transformation analysis, and the two-dimensional offset is mapped into the three-dimensional point cloud model of the ring wall surface by combining the pose parameter data set of each projector, so that the actual three-dimensional offset of each projection pixel can be obtained. After the offset is calculated, the offset is slightly compensated through bundle adjustment, three-dimensional geometric re-projection and least squares optimization method, the calibration landing point of each projection pixel on the three-dimensional point cloud model and the corresponding projection light direction are recalculated, and the calibration projection coordinates along the ring wall surface are generated. At this time, the user interaction logic previously generated ripple diffusion, particle sputtering or light effect enhancement and other interaction effects do not need to be recalculated, and only the original effective texture or particle position is remapped to the new projection coordinates based on the calibration projection coordinates.

[0116] In this way, the real-time interaction response data is determined by the output of the previous interaction logic and the current calibration coordinates, the content and interaction effect remain consistent, and the position can be completely aligned with the calibrated projection coordinates. This ensures that the projection picture displayed on the wall can respond to user operations in real time and ensure spatial accuracy. Finally, the calibrated projection coordinates and real-time interaction response data are input into the rendering engine to dynamically adjust the projection picture according to the environment and user operation. For example, there is a dynamic sea surface projection effect on the ring wall surface, and when the user clicks on the sea surface on the wall, the ripple diffusion interaction effect is triggered. When executing this interaction logic, the projection feedback image is continuously acquired. At a certain moment, due to the slight vibration of the projector, part of the projection area is offset downward by about 2 cm. The feedback image captures the actual landing point of the offset center position of the original ripple on the wall. The feature points in the feedback image are matched with the feature points of the original ripple texture, the offset is calculated, and the three-dimensional wall point cloud model is mapped by combining the projector pose parameters to obtain the calibration landing point of each pixel. Then, the projection light direction is fine-tuned to generate calibration projection coordinates. At this time, the original ripple particle data generated by the user's hand waving does not need to be recalculated, but the position of each ripple particle is remapped according to the calibration coordinates, so that the ripple is displayed at the correct position on the wall. Finally, when the next frame is rendered, the user still sees the ripple effect synchronized with the click action, but the projector offset has been corrected, and the real-time interaction response data records the new projection position, speed and direction information of each ripple particle, which is used to drive the rendering.

[0117] On the basis of the above technical solutions, the dynamic calibration and slight compensation processing based on the projection feedback image and the pose parameter data set is performed to generate the calibration projection coordinates along the ring wall surface and the corresponding real-time interaction response data, including: Obtaining each preset feature point corresponding to each projector and the theoretical coordinates of each preset feature point in the wall coordinate system, determining the actual pixel coordinates of each preset feature point in the projection feedback image based on the projection feedback image; Mapping the actual pixel coordinates to the wall coordinate system through coordinate system conversion to obtain the actual coordinates of each preset feature point in the wall coordinate system, and determining the pixel offset of the projection sub-picture corresponding to each projector based on the theoretical coordinates and the actual coordinates; Determining the actual contour of the projection sub-picture corresponding to each projector in the wall coordinate system based on the projection feedback image through coordinate system conversion, determining the theoretical contour of the projection sub-picture corresponding to each projector in the wall coordinate system based on the pose parameter data set, and performing geometric deformation analysis based on the actual contour and the theoretical contour to obtain the geometric deformation data of the projection sub-picture corresponding to each projector; Determining the actual overlapping area of each projection sub-picture based on the actual contour, determining the theoretical overlapping area of each projection sub-picture based on the theoretical contour, and determining the splicing error data of the projection sub-picture corresponding to each projector based on the actual overlapping area and the theoretical overlapping area; Based on the pixel offset, geometric deformation data and splicing error data, dynamic calibration and micro compensation processing are performed to generate calibration projection coordinates along the ring-shaped wall and corresponding real-time interactive response data.

[0118] In this scheme, the preset feature point is a reference point set in advance in the projection, usually selected at a key position of the projection picture, such as a corner point or a grid intersection point, used to calibrate the corresponding relationship between the projection and the wall.

[0119] The wall coordinate system is a three-dimensional coordinate system established with the ring-shaped wall as a reference, used to uniformly describe the positions of various points on the wall and the projection mapping relationship.

[0120] The theoretical coordinates are the ideal position coordinates of the preset feature points in the wall coordinate system, which can reflect the accurate positions on the wall where the projection picture should theoretically fall.

[0121] The actual pixel coordinates are measured through the projection feedback image, corresponding to the two-dimensional pixel position of each preset feature point in the camera or the projection feedback image.

[0122] The actual coordinates are three-dimensional coordinates obtained by mapping the actual pixel coordinates back to the wall coordinate system through coordinate system conversion, used for comparative analysis with the theoretical coordinates.

[0123] The projection sub-picture is an independent picture area projected on the wall by each projector, which is a component of the complete ring-shaped projection.

[0124] Pixel offset is the difference between actual and theoretical coordinates, which quantifies the displacement error of projected image on the wall.

[0125] Actual contour is the actual boundary shape of projected sub-image on the wall coordinate system, which is calculated by the feedback image and coordinate mapping.

[0126] Theoretical contour is the ideal boundary shape of projected sub-image on the wall coordinate system, which is calculated by the projector pose parameters and theoretical projection settings.

[0127] Geometric distortion data is the shape difference between actual and theoretical contours, which quantifies the geometric distortion of projected image on the wall.

[0128] Actual overlapping area is the actual overlapping part of multiple projected sub-images on the wall, which is used to evaluate the image stitching effect.

[0129] Theoretical overlapping area is the overlapping area that multiple projected sub-images should have under the theoretical settings, which is used to compare with the actual overlapping area.

[0130] Stitching error data is the difference between actual and theoretical overlapping areas, which quantifies the deviation and error of projected sub-image stitching.

[0131] A set of preset feature points is selected for each projector, which are mostly distributed in the key positions of the projected image, such as the corners, center point, and grid intersection. In the projection design stage, theoretical coordinates are set for these feature points in the wall coordinate system, which are the ideal landing positions of these points on the three-dimensional wall. These coordinates are calculated by measuring the size, shape of the wall, and the installation position of the projector, and then forming the theoretical projection mapping relationship. During actual projection, the projector projects the image onto the wall, while the camera or sensor synchronously captures the feedback image of the wall projection. With the help of corner detection, sub-pixel level edge extraction, or pattern matching image processing techniques, the two-dimensional actual pixel positions of each preset feature point are identified from the feedback image. After identification, the pixel coordinates are converted and mapped to the wall coordinate system through the calibration matrix to obtain the actual three-dimensional positions of the feature points on the wall. This conversion process fully considers the projector pose, wall curvature, and imaging distortion, and completes the coordinate conversion and geometric correction. The calibration matrix needs to be obtained in advance: first, collect the pixel coordinates of the calibration board or wall preset feature points with known dimensions in the projection feedback image, then pair them with the actual three-dimensional coordinates of these points in the wall coordinate system to establish the correspondence between pixel coordinates and world coordinates. Then use camera calibration algorithms such as Zhang Zhengyou calibration method or direct linear transformation method to calculate the calibration matrix containing focal length, principal point coordinates, and pixel scale, and combine distortion correction and minimum re-projection error optimization to ensure that the pixel coordinates can be accurately mapped to the three-dimensional coordinates of the wall.

[0132] After obtaining the actual coordinates, compare them with the preset theoretical coordinates to calculate the pixel offset of each projector, that is, the translation deviation and direction deviation between the actual position and the theoretical position of each feature point. These offsets can intuitively reflect the overall position deviation and local misalignment of the projection picture. On this basis, extract the projection sub-picture contour of each projector from the actual projection feedback image, and obtain the actual contour shape by mapping the picture edge or grid features to the three-dimensional wall coordinates. The theoretical contour is determined according to the projector installation pose parameters and projection design, that is, the position of the picture boundary projected by the projector in the ideal state on the wall coordinate system. By comparing the actual contour with the theoretical contour, the picture deformation can be analyzed to form geometric deformation data, which covers information such as stretching, compression, inclination or bending degree of the picture.

[0133] At the same time, the intersection of the actual contours of multiple projectors in the wall coordinate system is calculated to obtain the actual overlapping area; the theoretical overlapping area is obtained by calculating the intersection of the theoretical contours. By comparing the differences between the two, the splicing error data can be obtained, including overlapping deviation, insufficient overlap or excessive overlap, etc. Finally, based on the comprehensive pixel offset, geometric deformation data and splicing error data, each projection sub-picture is dynamically calibrated and finely compensated. Adjust the output position of the projection image, fine-tune the geometric mapping and color boundary, and generate calibrated projection coordinates that are continuous and seamlessly aligned along the ring-shaped wall. Fuse the calibration results with real-time user interaction information to generate real-time interactive response data for real-time rendering by the projector.

[0134] Step S105: Based on the calibrated projection coordinate mapping and real-time interactive response data, control each projector to perform projection display on the ring-shaped wall.

[0135] To accurately present the calibrated projection content on the ring-shaped wall, the calibration projection coordinates generated in the previous stage and real-time interactive response data are first received. According to the calibration projection coordinates, the pixels of the original rendering picture or texture map are mapped to the corresponding spatial positions on the three-dimensional point cloud of the wall, and the state of the particle system, wave field simulation or other dynamic effects is updated in combination with the real-time interactive response data, so that the position, intensity and direction of each effect element can be accurately corresponded to the calibration coordinates. Then the mapped texture and dynamic effect data are decomposed into the local projection coordinate system of each projector. According to the pose parameter data set, the projection light direction, focal length and projection angle of each projector are calculated, and the calibration coordinates on the three-dimensional point cloud of the wall are converted into two-dimensional projection plane coordinates of each projector. In this process, the position of each pixel or particle on the projector plane is accurately calculated through three-dimensional geometric re-projection, perspective transformation and ray tracing algorithm, to ensure that the pixels in the overlapping area of each projector are completely aligned without error, while the spatial continuity and visual coherence of the original dynamic effects are preserved. After the coordinate mapping is completed, the frame buffer of each projector is updated to a rendering frame containing the calibrated texture, particle data and interactive effect information. After receiving the frame buffer data, the projector accurately projects the pixels onto the ring-shaped wall through its own optical system, realizing the calibrated projection display. The entire process is continuously carried out at a fixed frame rate to ensure that each frame of projection picture is updated synchronously with the real-time interactive response data, so that each operation of the user on the wall produces visible dynamic feedback, adapting to environmental changes such as small vibration or optical offset of the projector.

[0136] Based on the above steps S101-S105, the projection picture can be highly consistent with the ring-shaped wall and user operation, while realizing dynamic calibration and real-time interaction, thereby improving the display accuracy and stability of the projection, and enhancing the immersion experience of the user.

[0137] Referring to the accompanying Figure 2 , Figure 2 is a main step flow diagram of a dynamic projection interactive calibration method of a ring-shaped science fiction space according to an embodiment of the present application. As shown in Figure 2 , the dynamic projection interactive calibration method of a ring-shaped science fiction space in the embodiment of the present application mainly includes the following steps S201-S207.

[0138] Step S201: Obtain three-dimensional point cloud data of a ring-shaped wall in a ring-shaped science fiction space, and obtain spatial pose information of projectors in the ring-shaped science fiction space, generate a ring-shaped wall point cloud model based on the three-dimensional point cloud data, and calibrate the projectors based on the ring-shaped wall point cloud model and the spatial pose information to obtain a pose parameter data set of the projectors.

[0139] Step S202: Based on the pose parameter dataset, control each projector to map the real-time state of each frame of pixels in the preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface.

[0140] Step S203: If user interaction data of the user on the ring-shaped wall surface is detected, determine an interaction area based on the user interaction data and execute corresponding logic.

[0141] Step S204: Continuously acquire projection feedback images of the ring-shaped wall surface in the process of executing the corresponding logic, perform dynamic calibration and micro compensation processing based on the projection feedback images and the pose parameter dataset, and generate calibrated projection coordinates along the ring-shaped wall surface and corresponding real-time interaction response data.

[0142] Step S205: Based on the calibrated projection coordinates mapping and the real-time interaction response data, control each projector to perform projection display on the ring-shaped wall surface.

[0143] Step S206: Continuously record the number of touches, the length of stay, and the sliding trajectory of the user on the ring-shaped wall surface within a preset time window, and generate user behavior characteristics based on the number of touches, the length of stay, and the sliding trajectory.

[0144] The preset time window is a continuous period of time determined in advance, which is used to count the touch actions of the user on the ring-shaped wall surface. All operation data of the user in this period of time will be recorded and analyzed.

[0145] The number of touches is the total number of times that the user actually touches the projection area of the ring-shaped wall surface within a set time.

[0146] The length of stay is the cumulative time that the user touches a certain place on the wall surface with a finger within a set time, which can directly reflect the degree of attention of the user to that place.

[0147] The sliding trajectory is the path that the user's finger traces when touching the wall surface, which is generally recorded as a series of coordinates. Based on these, the direction in which the user moves, the speed of movement, and the operation habits can be analyzed.

[0148] User behavior characteristics are descriptive information extracted from the original operation data of the number of touches, the length of stay, and the sliding trajectory, which are used to clearly describe the operation habits and preferences of the user.

[0149] The touch data collection function of the ring-shaped wall surface is opened at the start of the set time, and real-time monitoring of each touch of the user on the wall surface is performed. The time and three-dimensional position of the touch point at the time when the user's finger touches the wall surface are recorded, and the touch is calculated as a complete touch event, and the number of touches is increased. When the user lifts the hand to end the touch, the time of the touch is calculated, the end time is subtracted from the start time, the duration of the single stay is obtained, and the duration is added to the total stay time of the corresponding touch point or area. At the same time, if the user slides on the wall surface, the coordinates of the collection body touch are followed and the three-dimensional position at each time is recorded, and finally a complete sliding track is formed. The track can clearly show where the user moves, how fast, and how large the range touched.

[0150] During the entire set time, all recorded touch events are analyzed and sorted out to determine how many times each touch point is touched, how long each touch lasts, and what characteristics the trajectory of each slide has. Then, through statistical analysis, curvature estimation, time series feature extraction, trajectory clustering, and behavior pattern modeling algorithms, these raw data are converted into information that reflects user behavior, such as touch frequency to calculate the frequency of hand touch, total stay time to determine which area the user pays more attention to, and analysis of the curvature, speed change, and direction distribution of the trajectory to summarize the user's sliding habit pattern. Finally, a set of data is integrated to accurately reflect the user's operation strength, focus area, and operation habit. This entire collection and analysis process continues until the set time ends.

[0151] Step S207: updating the preset picture sequence based on the user behavior characteristics, and based on the pose parameter data set, controlling each projector to map the real-time state of each frame of pixels in the updated preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface.

[0152] After generating the user behavior characteristics, the preset picture sequence currently in use is first updated according to these characteristics. It carefully analyzes the recorded user behavior, such as the frequency of touch in different areas by the user operating the ring-shaped wall surface, the positions where the finger stays for a long time, the main direction of sliding, the change in speed, and the details of the trajectory curvature. In order to automatically adjust the picture according to the user's operation, these behavior information is converted into a set of picture adjustment weights with numerical values. For example, if a user touches a certain area frequently, the weight corresponding to that area will be increased; if the user always slides in a certain direction, the weight of the animation change related to that direction in the picture will also be increased.

[0153] After obtaining the adjustment weights, the preset picture sequence is reorganized and modified according to the adjustment weights. For example, if the user likes to slide to the right, more picture frames that spread to the right and extend to the right are selected, and a smoother or more smooth transition effect is added to the picture according to the sliding speed of the user. If the user stops in a certain area for a long time, the visual elements in the area can be highlighted, for example, enlarged and brightened. After screening, replacing, adding transitions or inserting new frames, the picture sequence is updated. After the picture is updated, each pixel is accurately projected onto the corresponding position of the ring-shaped wall. Therefore, the previously calibrated pose parameter data set is read first, which records the position, direction, projection angle and optical characteristics of each projector. Each pixel in the picture is processed one by one. First, the two-dimensional position of each pixel is determined, and then the optical parameters of the projector are combined to convert the pixel into a spatial light ray pointing to the wall. Then, the light ray is extended to the wall along the position and direction of the projector to calculate the landing point of the pixel on the ring-shaped wall.

[0154] Because the ring-shaped wall has an arc, the three-dimensional landing point is converted into a two-dimensional coordinate suitable for rendering on the ring-shaped wall in a circular arc expansion manner, for example, the horizontal position along the wall is converted into an angle description, and the vertical position is converted into a height description. In order to prevent the projected picture from being distorted, the calibration data obtained in the previous calibration is used to correct the coordinates, for example, to correct the offset error of the pixel in the projector, to correct the contour deformation caused by the arc of the wall, and to process the brightness in the overlapping area of multiple projectors to make the boundary of the spliced area not obvious. After the corrected coordinates of all pixels are calculated, accurate coordinate data that can be directly used for projection is obtained. The accurate coordinate data corresponds to the updated picture sequence one by one, generates complete projection instructions for each frame of picture, and makes all projectors display the updated dynamic picture on the ring-shaped wall synchronously. Finally, the picture can not only change in real time according to the touch, stop and sliding actions of the user, but also can be aligned strictly in space. The picture seen by the user is always natural and coherent, and the interaction is also strong.

[0155] In the embodiment, the behavior characteristics are generated by recording the touch frequency, stop duration and sliding trajectory of the user, the picture is dynamically updated and accurately projected onto the ring-shaped wall, and the content responds to the user interaction in real time.

[0156] On the basis of the above technical solution, after the method of controlling each projector to map the real-time state of each frame of pixel in the updated preset picture sequence to the corresponding projection coordinate on the ring-shaped wall, the method further comprises: Every time a preset update interval is reached, the touch frequency, stop duration and sliding trajectory of the user on the ring-shaped wall are continuously recorded in a preset time window, and the user behavior characteristics are generated based on the touch frequency, stop duration and sliding trajectory. renew the preset picture sequence based on the user behavior features, control each projector to map the real-time state of each frame of pixels in the updated preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface based on the pose parameter dataset.

[0157] In this scheme, the preset update interval is the time period for periodically reacquiring and analyzing user interaction data, that is, refreshing the statistics and feature extraction of user behavior once every certain time.

[0158] Each time the preset update interval is reached, the process of updating the preset picture sequence is triggered again, so as to continuously improve the immersion and experience fluency.

[0159] It should be noted that although the above embodiments describe the steps in a specific order, those skilled in the art can understand that, in order to achieve the effect of the present application, the different steps do not necessarily have to be executed in this order, they can be executed simultaneously (in parallel) or in other order, and these changes are within the protection scope of the present application.

[0160] Further, the present application also provides a dynamic projection interaction calibration system for a ring-shaped science fiction space.

[0161] Referring to the accompanying drawings Figure 3 , Figure 3 is the main structure block diagram of a dynamic projection interaction calibration system for a ring-shaped science fiction space according to an embodiment of the present application. As shown in Figure 3 , specifically includes: The calibration module 301 is configured to acquire three-dimensional point cloud data of the ring-shaped wall surface in the ring-shaped science fiction space, and acquire spatial pose information of each projector in the ring-shaped science fiction space, generate a ring-shaped wall surface point cloud model based on the three-dimensional point cloud data, and calibrate each projector based on the ring-shaped wall surface point cloud model and the spatial pose information to obtain a pose parameter dataset of each projector. The mapping module 302 is configured to control each projector to map the real-time state of each frame of pixels in the preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface based on the pose parameter dataset. The interaction module 303 is configured to determine an interaction region and execute corresponding logic based on the user interaction data if the user interaction data of the user on the ring-shaped wall surface is detected. The calibration module 304 is configured to continuously acquire a projection feedback image of the ring-shaped wall surface in the process of executing the corresponding logic, perform dynamic calibration and micro-compensation processing based on the projection feedback image and the pose parameter dataset, and generate a calibration projection coordinate along the ring-shaped wall surface and corresponding real-time interaction response data. The projection module 305 is configured to control the projectors to perform projection display on the ring-shaped wall surface based on the calibration projection coordinate mapping and the real-time interaction response data.

[0162] The dynamic projection interaction calibration system for a ring-shaped science fiction space provided by the embodiments of the present application can realize Figure 1 The method embodiments realize various processes, and thus details are not repeated here.

[0163] Those skilled in the art can understand that all or part of the processes in the method for implementing the above-mentioned embodiment can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned various method embodiments can be realized. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable storage medium can include any entity or device, medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium, etc. that can carry the computer program code. It should be noted that the computer readable storage medium can include or exclude some contents according to the requirements of legislation and patent practice in a jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer readable storage medium does not include electrical carrier signals and telecommunication signals.

[0164] Further, the present application also provides an electronic device 400, which includes a processor 401, a memory 402, a program or instruction stored in the memory 402 and executable on the processor 401. When the program or instruction is executed by the processor 401, various processes of the above-mentioned dynamic projection interaction calibration method for a ring-shaped science fiction space are realized, and the same technical effects can be achieved. To avoid repetition, details are not repeated here.

[0165] It should be noted that the electronic device in the embodiments of the present application includes the above-mentioned mobile electronic device and non-mobile electronic device.

[0166] Further, the present application also provides a computer readable storage medium. In an embodiment of the computer readable storage medium according to the present application, the computer readable storage medium can be configured to store a program of an interactive calibration method of dynamic projection of a ring-shaped science fiction space, which can be loaded and run by a processor to implement the above-mentioned interactive calibration method of dynamic projection of a ring-shaped science fiction space. For the convenience of illustration, only the parts related to the embodiments of the present application are shown, and the specific technical details not disclosed are referred to the method part of the embodiments of the present application. The computer readable storage medium can be a storage device formed by various electronic devices, and optionally, the computer readable storage medium in the embodiments of the present application is a non-transitory computer readable storage medium.

[0167] Obviously, the above-mentioned embodiments of the present application are only examples for clearly illustrating the present application, and are not intended to limit the embodiments of the present application. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, it is not necessary and impossible to enumerate all the embodiments. Any modification, equivalent replacement and improvement made within the spirit and principle of the present application shall be included in the protection scope of the claims of the present application.

Claims

1. A method for dynamic projection interaction calibration of a ring-shaped science fiction space, characterized in that, The method comprises: acquiring three-dimensional point cloud data of a ring-shaped wall surface in a ring-shaped science fiction space, and acquiring spatial pose information of each projector in the ring-shaped science fiction space, generating a ring-shaped wall surface point cloud model based on the three-dimensional point cloud data, and calibrating each projector based on the ring-shaped wall surface point cloud model and the spatial pose information to obtain a pose parameter data set of each projector; based on the pose parameter data set, controlling each projector to map the real-time state of each frame of pixels in a preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface; if user interaction data of a user on the ring-shaped wall surface is detected, determining an interaction area based on the user interaction data and executing corresponding logic; continuously acquiring projection feedback images of the ring-shaped wall surface during the execution of the corresponding logic, performing dynamic calibration and micro-compensation processing based on the projection feedback images and the pose parameter data set, generating calibration projection coordinates along the ring-shaped wall surface and corresponding real-time interaction response data; based on the calibration projection coordinates mapping and the real-time interaction response data, controlling each projector to project and display on the ring-shaped wall surface.

2. The method of claim 1, wherein, Wherein, if user interaction data of a user on the ring-shaped wall surface is detected, determining an interaction area based on the user interaction data and executing corresponding logic, comprising: if user interaction data of a user on the ring-shaped wall surface is detected, continuously extracting touch position data of the user based on the corresponding relationship between the user interaction data and the current projection picture of the ring-shaped wall surface within a continuous time; if the touch positions determined based on the touch position data within the continuous time all correspond to the water body interaction area of the current projection picture, continuously extracting touch force data based on the user interaction data; based on the touch position data, continuously determining a local neighborhood point set in the ring-shaped wall surface point cloud model centered on the touch position, and performing local surface fitting on the local neighborhood point set to obtain a local fitted surface corresponding to the touch position; based on the local fitted surface, continuously extracting the local surface characteristics and neighborhood topological structure of the touch position, and establishing a local coordinate system of the touch position based on the local surface characteristics and neighborhood topological structure, and continuously determining local geometric information of the touch position based on the local coordinate system; based on the local geometric information and the touch force data, dynamically configuring SPH simulation parameters, continuously acquiring current state data of the water body interaction system, and continuously updating the SPH simulation parameters based on the current state data; based on the continuously updated SPH simulation parameters, calculating the acceleration and motion trajectory of each floating interactive object in the local gravitational field of the water body interaction system, and based on the acceleration and motion trajectory, determining dynamic state data and real-time position data of each floating interactive object; based on the dynamic state data and real-time position data, calculating the dynamic projection coordinates of each floating interactive object on the ring-shaped wall surface, and controlling each projector to project and display on the ring-shaped wall surface based on the dynamic projection coordinates until no user interaction data of a user on the ring-shaped wall surface can be detected.

3. The method of claim 1, wherein, Wherein, if user interaction data of a user on the ring-shaped wall surface is detected, determining an interaction area based on the user interaction data and executing corresponding logic, further comprising: If the user interaction data of the user on the ring-shaped wall is detected, touch position data of the user is continuously extracted based on a correspondence relationship between the user interaction data and a current projection picture of the ring-shaped wall in continuous time; If the touch positions determined based on the touch position data in the continuous time all correspond to the starry sky interaction area of the current projection picture, touch force data and sliding direction data are continuously extracted based on the user interaction data; The touch force data is mapped into a rotating force field radius and a first rotating angular velocity by a nonlinear function, and a local rotating force field is generated in the starry sky interaction area with the touch position as the center based on the rotating force field radius and the first rotating angular velocity; Based on the touch position data and the rotating force field radius, lines located in the range of the local rotating force field are screened from a preset initial line set of the starry sky interaction area to obtain a target line set entering the local rotating force field; Based on the local rotating force field and the touch sliding direction, the rotating direction and a second rotating angular velocity of each line in the target line set are determined; Based on the rotating direction and the second rotating angular velocity, a rotating action is applied to each line, and a bending motion driven by original Perlin noise is superimposed to obtain a first dynamic motion state of each line; Based on the first dynamic motion state, a real-time motion speed of each line is calculated, and a display color of each line is determined based on the real-time motion speed and a preset color mapping rule; Based on the first dynamic motion state and the display color, real-time rendering data of each line is generated, and each projector is controlled to project and display on the ring-shaped wall based on the real-time rendering data until the user interaction data of the user on the ring-shaped wall cannot be detected.

4. The method of claim 1, wherein, Wherein, If the user interaction data of the user on the ring-shaped wall is detected, the interaction area is determined based on the user interaction data and the corresponding logic is executed, and the method further includes: If the user interaction data of the user on the ring-shaped wall is detected, touch position data of the user is continuously extracted based on a correspondence relationship between the user interaction data and a current projection picture of the ring-shaped wall in continuous time; If the touch positions determined based on the touch position data in the continuous time all correspond to the starry sky interaction area of the current projection picture, touch force data and sliding direction data are continuously extracted based on the user interaction data; The touch force data is mapped into a rotating force field radius and a first rotating angular velocity by a nonlinear function, and a local rotating force field is generated in the starry sky interaction area with the touch position as the center based on the rotating force field radius and the first rotating angular velocity; Based on the touch position data and the rotating force field radius, lines located in the range of the local rotating force field are screened from a preset initial line set of the starry sky interaction area to obtain a target line set entering the local rotating force field; Based on the local rotating force field and the touch sliding direction, the rotating direction and a second rotating angular velocity of each line in the target line set are determined; Based on the rotating direction and the second rotating angular velocity, a rotating action is applied to each line, and a bending motion driven by original Perlin noise is superimposed to obtain a first dynamic motion state of each line; Based on the first dynamic motion state, a real-time motion speed of each line is calculated, and a display color of each line is determined based on the real-time motion speed and a preset color mapping rule; Based on the first dynamic motion state and the display color, real-time rendering data of each line is generated, and each projector is controlled to project and display on the ring-shaped wall based on the real-time rendering data until the user interaction data of the user on the ring-shaped wall cannot be detected. Wherein, If the user interaction data of the user on the ring-shaped wall is detected, the interaction area is determined based on the user interaction data and the corresponding logic is executed, and the method further includes: If the user interaction data of the user on the ring-shaped wall is detected, touch position data of the user is continuously extracted based on a correspondence relationship between the user interaction data and a current projection picture of the ring-shaped wall in continuous time; If the touch positions determined based on the touch position data in the continuous time all correspond to the starry sky interaction area of the current projection picture, touch force data and sliding direction data are continuously extracted based on the user interaction data; The touch force data is mapped into a rotating force field radius and a first rotating angular velocity by a nonlinear function, and a local rotating force field is generated in the starry sky interaction area with the touch position as the center based on the rotating force field radius and the first rotating angular velocity; Based on the touch position data and the rotating force field radius, lines located in the range of the local rotating force field are screened from a preset initial line set of the starry sky interaction area to obtain a target line set entering the local rotating force field; Based on the local rotating force field and the touch sliding direction, the rotating direction and a second rotating angular velocity of each line in the target line set are determined; Based on the rotating direction and the second rotating angular velocity, a rotating action is applied to each line, and a bending motion driven by original Perlin noise is superimposed to obtain a first dynamic motion state of each line; Based on the first dynamic motion state, a real-time motion speed of each line is calculated, and a display color of each line is determined based on the real-time motion speed and a preset color mapping rule; Based on the first dynamic motion state and the display color, real-time rendering data of each line is generated, and each projector is controlled to project and display on the ring-shaped wall based on the real-time rendering data until the user interaction data of the user on the ring-shaped wall cannot be detected. Perform position update on each associated text node based on the second dynamic motion state to obtain text node update result after elastic response, and control each projector to perform real-time rendering display based on the text node update result until no user interaction data of the user on the ring-shaped wall surface can be detected.

5. The method of claim 1, wherein, In the method, Perform dynamic calibration and micro-compensation processing based on the projection feedback image and the pose parameter data set to generate calibration projection coordinates along the ring-shaped wall surface and corresponding real-time interaction response data, including: Obtain each preset feature point corresponding to each projector and theoretical coordinates of each preset feature point in a wall surface coordinate system, determine actual pixel coordinates of each preset feature point in the projection feedback image based on the projection feedback image; Map the actual pixel coordinates to the wall surface coordinate system through coordinate system conversion to obtain actual coordinates of each preset feature point in the wall surface coordinate system, and determine pixel offset of a projection sub-picture corresponding to each projector based on the theoretical coordinates and the actual coordinates; Determine actual outlines of the projection sub-picture corresponding to each projector in the wall surface coordinate system based on the projection feedback image through coordinate system conversion, determine theoretical outlines of the projection sub-picture corresponding to each projector in the wall surface coordinate system based on the pose parameter data set, and perform geometric deformation analysis based on the actual outlines and the theoretical outlines to obtain geometric deformation data of the projection sub-picture corresponding to each projector; Determine actual overlapping areas of each projection sub-picture based on the actual outlines, determine theoretical overlapping areas of each projection sub-picture based on the theoretical outlines, and determine splicing error data of the projection sub-picture corresponding to each projector based on the actual overlapping areas and the theoretical overlapping areas; Perform dynamic calibration and micro-compensation processing based on the pixel offset, the geometric deformation data, and the splicing error data to generate calibration projection coordinates along the ring-shaped wall surface and corresponding real-time interaction response data.

6. The method of claim 1, wherein, In the method, After controlling each projector to perform projection display on the ring-shaped wall surface based on the calibration projection coordinate mapping and the real-time interaction response data, the method further includes: Continuously record the number of touches, the length of stay, and the sliding track of the user on the ring-shaped wall surface within a preset time window, and generate user behavior features based on the number of touches, the length of stay, and the sliding track; Update the preset picture sequence based on the user behavior features, and control each projector to map the real-time state of each frame of pixels in the updated preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface based on the pose parameter data set.

7. The method of claim 6, wherein, In the method, After controlling each projector to map the real-time state of each frame of pixels in the updated preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface, the method further includes: Re-continuously record the number of touches, the length of stay, and the sliding track of the user on the ring-shaped wall surface within a preset time window every time a preset update interval is reached, and generate user behavior features based on the number of touches, the length of stay, and the sliding track; Update the preset picture sequence based on the user behavior features, and control each projector to map the real-time state of each frame of pixels in the updated preset picture sequence to the corresponding projection coordinates on the ring-shaped wall surface based on the pose parameter data set.

8. A dynamic projection interactive calibration system for a circular science fiction space, characterized by, The system includes: The calibration module is configured to continuously acquire a projection feedback image of the annular wall surface during execution of the corresponding logic, perform dynamic calibration and micro-compensation processing based on the projection feedback image and the pose parameter data set, generate a calibration projection coordinate along the annular wall surface, and generate corresponding real-time interaction response data. The mapping module is configured to control each of the projectors to map a real-time state of each frame of pixels in a preset picture sequence to a corresponding projection coordinate on the annular wall surface based on the pose parameter data set. The interaction module is configured to determine an interaction region and execute corresponding logic based on user interaction data of a user on the annular wall surface if the user interaction data is detected. The calibration module is configured to continuously acquire a projection feedback image of the annular wall surface during execution of the corresponding logic, perform dynamic calibration and micro-compensation processing based on the projection feedback image and the pose parameter data set, generate a calibration projection coordinate along the annular wall surface, and generate corresponding real-time interaction response data. The projection module is configured to control each of the projectors to perform projection display on the annular wall surface based on the calibration projection coordinate mapping and the real-time interaction response data.

9. An electronic device comprising a processor, a memory, and a program or instructions stored on the memory and executable on the processor, characterized in that, The program or instructions are adapted to be loaded and run by the processor to execute the dynamic projection interaction calibration method of the annular science fiction space according to any one of claims 1 to 7.

10. A computer readable storage medium having stored therein a plurality of program codes, characterized in that, The program code is adapted to be loaded and run by the processor to execute the dynamic projection interaction calibration method of the annular science fiction space according to any one of claims 1 to 7.

Images