Art Design Virtual Reality Content Synchronous Display System Based on AI and Image Detection

Abstract

This invention relates to the field of virtual reality display technology, specifically to a synchronized display system for art design virtual reality content based on AI and image detection. The system includes: a feature acquisition unit that extracts an art design feature set, including brushstroke trajectory features, material reflectivity features, and spatial topological relationship features; a parameter matching unit that acquires display characteristic parameters of the target VR terminal and generates a sequence of virtual reality content frames that match these parameters; and a virtual reality generation unit that feeds the updated art design feature set back to the generative adversarial network model for incremental content rendering and pushes the adapted virtual reality content to multiple terminals via a distributed synchronization engine. This invention estimates the feature update amount based on changes in trigger pressure in the user's real-time interaction trajectory, ensuring that incremental content rendering is consistent with the user's dynamic operations, thus solving the problems of large response latency and poor device adaptation in existing systems.

Description

Technical Field

[0001] This invention relates to the field of virtual reality display technology, and in particular to a synchronized display system for art-designed virtual reality content based on AI and image detection. Background Technology

[0002] With the rapid development of virtual reality (VR) technology, the art and design field is gradually exploring the transfer of visual elements such as two-dimensional painting, three-dimensional modeling, and material textures to virtual space for immersive display. However, existing digital display systems for art and design generally suffer from the following technical bottlenecks:

[0003] First, traditional image acquisition methods primarily rely on RGB three-channel information, making it difficult to capture multi-dimensional features such as surface reflection, material layers, and microscopic topological structures. This results in significant errors in virtual models' ability to reproduce realistic artistic styles and a lack of faithful representation of detailed textures. Second, most existing content generation mechanisms employ static texture mapping or pre-rendered models, lacking the ability to interact with user behavior. This leads to issues such as high latency and sluggish responses in dynamic responses, haptic feedback, and personalized adjustments.

[0004] Furthermore, in content delivery scenarios targeting multi-terminal VR devices, current systems generally adopt a unified rendering content broadcasting or simple resolution compression method, which is difficult to adapt to the display characteristics of different terminals (such as refresh rate, field of view, color gamut coverage), and is prone to image tearing, rendering distortion and synchronization lag, which seriously affects the user's immersion and interactive experience. Summary of the Invention

[0005] This invention provides an art design virtual reality content synchronous display system based on AI and image detection. It can integrate high-dimensional image detection, AI feature modeling and multi-terminal adaptive art design virtual reality content display system to achieve realistic restoration of art entities, interactive content updates and low-latency multi-device synchronous push.

[0006] An AI- and image detection-based virtual reality content synchronization display system for art design includes:

[0007] Feature acquisition unit: acquires surface texture data of the art design entity object through a multispectral image acquisition device, and extracts at least three art design feature sets using a dynamic semantic segmentation network. The art design feature sets include brush stroke trajectory features, material reflectivity features, and spatial topological relationship features.

[0008] Parameter matching unit: Inputs the art design feature set into the pre-trained generative adversarial network model, synchronously obtains the display characteristic parameters of the target VR terminal through device parameter parsing, and generates a virtual reality content frame sequence that matches the display characteristic parameters. The virtual reality content frame sequence includes a dynamic light and shadow effect layer and interactive response event markers.

[0009] Virtual Reality Generation Unit: Monitors spatial pose data generated by user interaction in real time, triggers feature update instructions based on the interaction response event markers, feeds back the updated art design feature set to the generative adversarial network model for incremental content rendering, and pushes the adapted virtual reality content to multiple terminals through a distributed synchronization engine.

[0010] Optionally, in the feature acquisition unit, based on a multispectral image acquisition device, the surface of the artistic design object is scanned three times in an orthogonal polarization mode within a preset wavelength range to acquire visible light reflectance, near-infrared absorption, and short-wave infrared scattering spectra, respectively.

[0011] Optionally, the feature acquisition unit further includes constructing a dual-stream dynamic semantic segmentation network, wherein the first sub-network uses dilated convolution kernels to extract the spatiotemporal continuity features of the pen stroke trajectory and tracks the pen stroke pressure change curve through a long short-term memory module; the second sub-network uses a spectral feature fusion layer to weight and stitch together the channel dimensions of the three sets of maps, and outputs a pixel-level feature map including the material reflectance gradient value.

[0012] Optionally, the dual-stream dynamic semantic segmentation network deploys a three-dimensional graph attention mechanism at its end. By establishing a curvature change correlation matrix between adjacent pixels, it generates non-Euclidean feature vectors representing spatial topological relationships. Finally, the spatiotemporal continuity features, pixel-level feature maps, and non-Euclidean feature vectors are combined into the art design feature set.

[0013] Optionally, the generative adversarial network model in the parameter matching unit adopts a multimodal conditional generative adversarial network, including a generator G and a discriminator D.

[0014] Optionally, the input of the generator G is connected to three parallel channels of the art design feature set:

[0015] The first channel maps the pen stroke trajectory features to a latent space vector, and generates a basic geometric mesh through a spatiotemporal encoder;

[0016] The second channel inputs the material reflectivity characteristics into the physical rendering pipeline, and combines them with the Monte Carlo ray tracing algorithm to generate dynamic lighting and shadow effect layers.

[0017] The third channel receives the display characteristic parameter set output by the device parameter parsing, including the target VR terminal's screen refresh rate, color gamut coverage, and field of view curvature radius. The parameter adaptation layer then generates device-related content rendering constraints.

[0018] Optionally, the discriminator D is equipped with a device perception verification mechanism, specifically including:

[0019] Color gamut matching verification: Calculate the ΔE2000 color difference between the generated content and the target device's P3 color gamut;

[0020] Motion blur prediction: Predicts the image ghosting index based on screen refresh rate and eye-tracking data;

[0021] Geometric distortion correction: Dynamically adjust the deformation compensation parameters of the vertex shader based on the field of view curvature radius.

[0022] Optionally, the method further includes embedding interactive response event markers in the content frame sequence via an event marker generator, the event markers including:

[0023] Haptic feedback trigger area: Pressure response levels are divided according to the material reflectivity gradient value;

[0024] Light and shadow interaction sensitive area: Dynamic shadow update priority is marked based on ray tracing results;

[0025] Device compatibility identifier: Records the hash value of rendering constraints for verification by the distributed synchronization engine.

[0026] Optionally, the virtual reality generation unit specifically includes:

[0027] a) The six-degree-of-freedom pose data is obtained by fusing a nine-axis inertial measurement unit with an infrared optical tracking system, and a filter is used to eliminate jitter noise to generate a clean pose stream including timestamps.

[0028] b. Construct an event-marking driven engine to trigger feature update operations based on user interaction trigger areas:

[0029] c employs differential coding technology in the incremental rendering stage:

[0030] The feature hash comparison algorithm identifies the regions that need to be updated, and only the feature subsets with a hash value difference greater than 15% are regenerated;

[0031] A progressive generation strategy is used: the first frame generates a low-resolution base mesh, and subsequent frames are gradually improved to the target resolution through residual connections;

[0032] When the distributed synchronization engine performs multi-terminal push:

[0033] Attach a device fingerprint identifier to each content frame, including screen color temperature calibration value and rendering pipeline latency compensation coefficient;

[0034] A layered synchronization protocol is adopted: the base layer transmits geometric topology data, and the enhanced laminar transmission transmits lighting and shadow effects data;

[0035] When version differences between terminals are detected, a content merging algorithm based on the minimum spanning tree is triggered.

[0036] Optionally, the event tagging engine execution logic is as follows:

[0037] Analyze the boundary coordinates of the haptic feedback trigger area in the interactive response event markers;

[0038] When the operation trajectory in the cleanup pose flow intersects with the trigger region:

[0039] Calculate the feature set update amount ΔF based on the pressure-displacement curve;

[0040] Generate update instruction packages with priority tags, the priority of which is determined by the heat value Hv of the interactive area.

[0041] The beneficial effects of this invention are:

[0042] This invention combines multispectral image acquisition with a dual-stream dynamic semantic segmentation network to heterogeneously decouple and jointly model brushstroke trajectories, material reflectivity, and spatial topological relationships. In particular, it significantly improves the signal-to-noise ratio of feature extraction by using orthogonal polarization imaging under high reflectivity interference conditions. The non-Euclidean spatial vectors generated by the graph attention mechanism can fully restore the three-dimensional configuration of complex artistic textures, solving the problem of lack of material physical semantics in traditional RGB image extraction, and providing a higher-precision digital restoration foundation for virtual artwork modeling.

[0043] This invention constructs a multimodal conditional generative adversarial network, deeply integrating artistic feature sets with VR terminal display parameters. It introduces three discrimination mechanisms: color gamut matching verification, motion blur prediction, and geometric distortion correction, enabling fine-grained adaptation and adjustment of content frames. Simultaneously, by leveraging a haptic response event-driven engine and an LSTM structure, it estimates the feature update amount of trigger pressure changes in the user's real-time interaction trajectory, ensuring that incremental content rendering is consistent with the dynamics of user operation, thus solving the problems of large response latency and poor device adaptation in existing systems.

[0044] This invention proposes a differential encoding mechanism based on hash comparison and residual connection, combined with a progressive resolution enhancement strategy, to perform target-level rendering only on sub-regions in the content that have undergone significant changes, thereby reducing bandwidth and computing power consumption. Furthermore, by combining device fingerprint identification and minimum spanning tree content merging algorithm, a multi-terminal push system that balances low latency and consistency is constructed, effectively solving problems such as difficult synchronization and slow merging in heterogeneous rendering scenarios of multiple devices, and ensuring the overall optimal performance, quality, and latency of the virtual reality experience. Attached Figure Description

[0045] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 This is a schematic diagram of the system functional units according to an embodiment of the present invention;

[0047] Figure 2 This is a schematic diagram of the device perception verification mechanism according to an embodiment of the present invention. Detailed Implementation

[0048] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. It should also be noted that, to make the embodiments more comprehensive, the following embodiments are the best and preferred embodiments, and those skilled in the art can use other alternative methods to implement some well-known technologies; moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0049] like Figures 1-2 As shown, the art design virtual reality content synchronous display system based on AI and image detection includes:

[0050] Feature acquisition unit: Acquires surface texture data of art design entity objects through a multispectral image acquisition device, and extracts at least three art design feature sets using a dynamic semantic segmentation network. The art design feature sets include brush stroke trajectory features, material reflectivity features, and spatial topological relationship features.

[0051] Parameter matching unit: Inputs the art design feature set into the pre-trained generative adversarial network model, synchronously obtains the display characteristic parameters of the target VR terminal through device parameter parsing, and generates a virtual reality content frame sequence that matches the display characteristic parameters. The virtual reality content frame sequence includes a dynamic lighting effect layer and interactive response event markers.

[0052] Virtual Reality Generation Unit: It monitors the spatial pose data generated by user interaction in real time, triggers feature update instructions based on interaction response event markers, feeds the updated art design feature set back to the generative adversarial network model for incremental content rendering, and pushes the adapted virtual reality content to multiple terminals through a distributed synchronization engine.

[0053] The feature acquisition unit specifically includes:

[0054] Multispectral image collaborative acquisition subunit: Employing a multispectral imaging module with a wavelength range of 380-2500nm, the surface of the artistic design object is scanned three times in orthogonal polarization mode, acquiring the following spectral data respectively:

[0055] Visible light reflectance spectrum: λ1∈[400nm,700nm];

[0056] Near-infrared absorption spectrum: λ2∈[701nm,1100nm];

[0057] Short-wave infrared scattering spectrum: λ3∈[1101nm,2500nm];

[0058] Orthogonal polarization is used to suppress specular reflection interference and improve the signal-to-noise ratio by 60dB; the spectral resolution is controlled with an accuracy of ±5nm to resolve the structure of subsurface materials.

[0059] Semantic segmentation network construction sub-unit: Constructing a two-stream neural network structure consisting of two sub-networks:

[0060] The first sub-network is used to extract the spatiotemporal continuity features of the pen stroke trajectory. The network structure is configured as follows:

[0061] The convolution module uses a dilated convolution kernel, and the dilation coefficient is set to:

[0062] d∈{2,4,8},stride=2,receptive field=7×7, where d is the dilation coefficient of the dilated convolution kernel, stride is the stride of the convolution kernel, and receptive field is the size of the receptive field;

[0063] An LSTM module was subsequently added for dynamic tracking, configured as follows:

[0064] Hidden layer dimension = 256, time window length = 15 frames;

[0065] The second sub-network processes spectral data from the three sets of maps, and uses a spectral feature fusion layer to perform channel concatenation and weighting, outputting a pixel-level feature map containing the material reflectance gradient. The fusion calculation method is as follows:

[0066] Among them, Ffused This represents a pixel-level material feature map (reflectance gradient map) obtained by weighted fusion of three sets of spectral images. For the feature map corresponding to the i-th band spectrum, w i This represents a weight factor that is dynamically adjusted through a learnable parameter matrix, satisfying ∑w i =1.

[0067] Relationship modeling and feature set combination subunit: A 3D graph attention mechanism is introduced at the end of the two-stream network to generate non-Euclidean space feature vectors.

[0068] Establish the curvature variation correlation matrix between pixels:

[0069] Where K(i,j) represents the Gaussian curvature of the i,j-th pixel, and z(x,y) represents the grayscale height or reflectance value of that pixel. This represents the second partial derivative of z with respect to the x-direction, i.e., the rate of change of concavity / convexity along the horizontal direction. Let z represent the second partial derivative of z with respect to y, i.e. the rate of change of concavity and convexity along the vertical direction, and (i,j) represent the pixel coordinates of the i-th row and j-th column in the image.

[0070] A graph attention mechanism is introduced, employing eight parallel attention heads for non-Euclidean encoding, ultimately combining the three types of features to form a complete art and design feature set:

[0071]

[0072] in, This indicates the spatiotemporal continuity of the brushstroke trajectory. This represents a pixel-level material reflectivity gradient feature map. Let R represent the non-Euclidean space topological vector obtained by mapping hyperbolic geometric space, with a mapping radius of curvature R = 0.8.

[0073] The LSTM module is used to model the temporal changes in pen stroke trajectory, especially the trend of pen pressure changing over time. The following is the state calculation process for each time step t:

[0074] 1. Input gate: Controls the current input x t Whether the cell status has been written:

[0075] i t =σ(W i x t +U i h t-1 +b i );

[0076] 2. Forget Gate: Controls the state c of the previous step. t-1Whether to retain: f t =σ(W f x t +U f h t-1 +b f );

[0077] 3. Output gate: Control unit state c t Which parts are used in the current output:

[0078] o t =σ(W o x t +U o h t-1 +b o );

[0079] 4. Candidate State: Generate candidate memory content for the current time step:

[0080]

[0081] 5. Unit State Update: Update the current memory state based on the forget gate and the input gate:

[0082]

[0083] 6. Current Output: Calculate the current hidden state by combining the output gate and the current memory state:

[0084] h t =o t ⊙tanh(c t );

[0085] Above: x t Let x be the current input vector. t It is the feature vector of the brush stroke image extracted from frame t, including the spatial location information of the brush stroke path and the corresponding pressure value encoding, used to reflect the local spatiotemporal state of the drawing action in that frame, h t-1 c is the output of the previous time step. t-1 i represents the cell state of the previous time step. t ,f t ,o t These are the input gate, forget gate, and output gate, respectively. The candidate cell states are at the current time step, σ(·) is the sigmoid activation function, tanh(·) is the hyperbolic tangent activation function, and ⊙ represents element-wise multiplication; W * U * ,b * These are the trainable weight matrix and bias parameters.

[0086] The LSTM module operates on the time-series input {x1,x2,…,x} of the pen stroke trajectory features. n The output sequence {h1,h2,…,h} is calculated recursively. n} is used to represent the pressure change trend during continuous drawing and is embedded into the feature set as a high-order dynamic feature of the stroke trajectory. middle.

[0087] The parameter matching unit specifically includes:

[0088] Multimodal Conditional Generative Adversarial Network (MC-GAN) Construction Subunit: Construct a multimodal conditional generative adversarial network (MC-GAN) whose generator G receives three types of parallel inputs from an art and design feature set:

[0089] First channel: Characterizing brushstroke trajectory features Mapped to latent space vectors, the underlying geometric mesh is generated using a spatiotemporal encoder: ; where CapsEnc(·) represents an encoder using a capsule network, with the capsule network dimensions set to [8,16,32];

[0090] Second channel: Utilizing material reflectivity characteristics The input is fed into the physical rendering pipeline, and a dynamic lighting and shadow effect layer is generated using the Monte Carlo ray tracing algorithm:

[0091] N = 256; where I(x,y) is the final synthesized image brightness value at pixel (x,y), and L i (x,y) represents the lighting result of the i-th path, N is the number of sampling points per pixel, and the depth is set to 8. This means that after each light reflection, the path is terminated based on probability to avoid all paths bouncing indefinitely and reduce the amount of computation. The "depth of 8" means that the light is allowed to bounce a maximum of 8 times. Paths exceeding this depth will be forcibly terminated.

[0092] Third channel: Receives target device parameter set Includes the following:

[0093] Among them, f VR For screen refresh rate, range: 90–144, C color R represents color gamut coverage (DCI-P3). fov The radius of curvature of the field of view ranges from 500 to 1500 mm.

[0094] The parameter adaptation layer adjusts the resolution of the output image, and the interpolation calculation error ∈ res Control is: ∈ res <0.5 pixels, ∈ resThis indicates the interpolation deviation between the content image and the device resolution.

[0095] The device awareness verification subunit of the discriminator: In the adversarial network discriminator D, a device adaptation awareness verification mechanism is integrated.

[0096] Color gamut matching verification: Calculated based on the CIE ΔE2000 color difference formula, with the matching standard being: ΔE 00 <2.3, the calculation parameters follow: 2° viewing angle, standard light source D65 conditions, ΔE 00 The CIE 2000 color difference value is used to measure the color deviation between the generated image and the target device's standard color gamut (P3).

[0097] Motion blur prediction: Predicting the ghosting index of generated content on VR devices Represented as:

[0098] Where τ is the response time constant, t is the response time, and v max For the maximum eye movement angular velocity, f VR For refresh rate, the ghosting threshold is I<5%.

[0099] Geometric distortion correction: based on the radius of curvature R of the field of view. fov The distortion compensation for the vertex shader is performed using the following distortion parameters: k1∈[-0.15,0.15], p1,p2∈[-0.05,0.05], where k1 is the radial distortion coefficient and p1,p2 are the tangential distortion coefficients.

[0100] Interactive Response Event Tag Embedding Subunit: Embeds response regions into the output frame sequence using an event tag generator, specifically including:

[0101] Haptic feedback trigger area: based on material reflectivity gradient value Classification of tactile response levels:

[0102] F p ∈[0.1,5]N / mm 2 F p This is the tactile feedback pressure response value, reflecting the required tactile feedback intensity for different areas;

[0103] The grading is based on the Hertzian contact model, with a mesh accuracy of 0.1 mm².

[0104] Light and shadow interaction sensitive area: When the change rate of the shadow edge satisfies: ΔS / Δt>15% / frame, it is marked as a high priority update area. ΔS is the amount of pixel area change of the shadow edge, which represents the amount of change of the shadow shape between two frames. Δt is the time interval between adjacent frames, which is equal to 1.

[0105] Device adaptation identifier: Generates hash values ​​from rendering constraints for distributed synchronous verification.

[0106] The hash generation and verification delay T is among them. verify Controlled at: T verify <0.3ms.

[0107] The virtual reality generation unit specifically includes:

[0108] (a) Six-DOF Pose Fusion and Cleaning: A nine-axis inertial measurement unit (IMU) and an infrared optical tracking system are used to fuse six-DOF pose data (positions x, y, z and attitude angles φ, θ, ψ), and an extended Kalman filter (EKF) is used to eliminate noise and generate a cleaned pose flow.

[0109] Pose calculation frequency: f pose ≥120Hz;

[0110] Attitude angle error: ∈ θ <0.1°;

[0111] Spatial positioning standard deviation: σ x,y,z ≤0.3mm;

[0112] The Kalman filter noise covariance matrix is ​​set as follows:

[0113] Q=diag([0.01,0.01,0.01,0.001,0.001,0.001]);

[0114] The IMU sensor parameter range is:

[0115] Acceleration range: ±16g; Gyroscope range: ±2000° / s; Magnetometer range: ±4900μT.

[0116] (b) Event-driven engine: Triggers feature update operations based on user interaction trigger areas.

[0117] Feature set update calculation: When the operation trajectory intersects with the haptic feedback area, the update amount is calculated as follows:

[0118] ΔF=K p ·∫(F(t)-F0)dt; where ΔF is the characteristic adjustment amount to be injected, K p F(t) is the material stiffness coefficient (0.8 for canvas and 0.3 for watercolor paper), F(t) is the pressure at the current interaction moment, and F0 is the trigger threshold reference pressure.

[0119] The interaction popularity value is calculated as follows: Among them, H v N represents the heat value of the interactive area. interactThe number of interactions within a unit time window. The pressure gradient (intensity of regional pressure distribution change) is represented by T, which is the time window. The sliding window mechanism has a length of 500 ms and a step size of 100 ms. α = 0.6 and β = 0.4 represent the weighting coefficients.

[0120] (c) Incremental rendering and differential coding: Rendering optimization employs differential coding and progressive generation strategies.

[0121] Hash comparison strategy: Use CityHash64 to perform a comparison on a subset of features in each frame. If the hash difference ΔH > 15%, then trigger the regeneration of that region.

[0122] Asymptotic residual generation:

[0123] The basic grid size for the first frame output is: R0 = 256 × 256;

[0124] Subsequent stepwise residual upsampling: R i =2 i • 256, i = 1, 2, ..., 3 (up to 2048 × 2048), and a deformable convolution kernel is introduced in each upsampling stage to adapt to local structural changes.

[0125] (d) Distributed synchronization and content push strategy: Multi-terminal synchronization relies on device identification and hierarchical data stream structure.

[0126] Frame Attached Device Fingerprint: Among them, T white This is the screen color temperature calibration value, ranging from 2000-10000K, λ delay The rendering pipeline delay compensation factor ranges from [0.8, 1.2]. The device fingerprint is transmitted using HMAC-SHA1 encryption.

[0127] Layered synchronization protocol:

[0128] The base layer transmits geometric topology data, with a frame data size ≤ 5KB / frame;

[0129] Enhancement layer transmits light and shadow effects streams: bitrate ∈ [8,50] Mbps;

[0130] Content merging mechanism: When content version differences are detected between terminals, a content merging algorithm based on minimum spanning tree (MST) is used.

[0131] Network latency weight w i,j Set as:

[0132] Merging processing delay τ merge Controlled at: τ merge <10ms.

[0133] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0134] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A synchronized display system for virtual reality content of art design based on AI and image detection, characterized in that, include: Feature acquisition unit: acquires surface texture data of the art design entity object through a multispectral image acquisition device, and extracts at least three art design feature sets using a dynamic semantic segmentation network. The art design feature sets include brush stroke trajectory features, material reflectivity features, and spatial topological relationship features. Parameter matching unit: Inputs the art design feature set into the pre-trained generative adversarial network model, synchronously obtains the display characteristic parameters of the target VR terminal through device parameter parsing, and generates a virtual reality content frame sequence that matches the display characteristic parameters. The virtual reality content frame sequence includes a dynamic light and shadow effect layer and interactive response event markers. The generative adversarial network model adopts a multimodal conditional generative adversarial network, including a generator G and a discriminator D; The input of the generator G is connected to three parallel channels of the art design feature set: The first channel maps the pen stroke trajectory features to a latent space vector, and generates a basic geometric mesh through a spatiotemporal encoder; The second channel inputs the material reflectivity characteristics into the physical rendering pipeline, and combines them with the Monte Carlo ray tracing algorithm to generate dynamic lighting and shadow effect layers. The third channel receives the display characteristic parameter set output by the device parameter parsing, including the screen refresh rate, color gamut coverage and field of view curvature radius of the target VR terminal, and generates device-related content rendering constraints through the parameter adaptation layer; Virtual Reality Generation Unit: Monitors spatial pose data generated by user interaction in real time, triggers feature update instructions based on the interaction response event markers, feeds back the updated art design feature set to the generative adversarial network model for incremental content rendering, and pushes the adapted virtual reality content to multiple terminals through a distributed synchronization engine.

2. The AI- and image detection-based art design virtual reality content synchronous display system according to claim 1, characterized in that, In the feature acquisition unit, based on the multispectral image acquisition device, the surface of the artistic design object is scanned three times in an orthogonal polarization mode within a preset wavelength range to acquire visible light reflectance spectrum, near-infrared absorption spectrum and short-wave infrared scattering spectrum respectively.

3. The AI- and image detection-based art design virtual reality content synchronous display system according to claim 2, characterized in that, The feature acquisition unit also includes a dual-stream dynamic semantic segmentation network. The first sub-network uses dilated convolution kernels to extract the spatiotemporal continuity features of the pen stroke trajectory and tracks the pen stroke pressure change curve through a long short-term memory module. The second sub-network uses a spectral feature fusion layer to weight and stitch together the channel dimensions of the three sets of maps, and outputs a pixel-level feature map including the material reflectance gradient value.

4. The AI- and image detection-based art design virtual reality content synchronous display system according to claim 3, characterized in that, The dual-stream dynamic semantic segmentation network deploys a 3D graph attention mechanism at its end. By establishing a curvature change correlation matrix between adjacent pixels, it generates non-Euclidean feature vectors that characterize spatial topological relationships. Finally, it combines the spatiotemporal continuity features, pixel-level feature maps, and non-Euclidean feature vectors into the art design feature set.

5. The AI- and image detection-based art design virtual reality content synchronous display system according to claim 1, characterized in that, The discriminator D is equipped with a device sensing and verification mechanism, specifically including: Color gamut matching verification: Calculate the ΔE2000 color difference between the generated content and the target device's P3 color gamut; Motion blur prediction: Predicts the image ghosting index based on screen refresh rate and eye-tracking data; Geometric distortion correction: Dynamically adjust the deformation compensation parameters of the vertex shader based on the field of view curvature radius.

6. The AI- and image detection-based art design virtual reality content synchronous display system according to claim 5, characterized in that, The parameter matching unit further includes embedding interactive response event markers in the content frame sequence through an event marker generator, the event markers including: Haptic feedback trigger area: Pressure response levels are divided according to the material reflectivity gradient value; Light and shadow interaction sensitive area: Dynamic shadow update priority is marked based on ray tracing results; Device compatibility identifier: Records the hash value of rendering constraints for verification by the distributed synchronization engine.

7. The AI- and image detection-based art design virtual reality content synchronous display system according to claim 1, characterized in that, The virtual reality generation unit specifically includes: a) The six-degree-of-freedom pose data is obtained by fusing a nine-axis inertial measurement unit with an infrared optical tracking system, and a filter is used to eliminate jitter noise to generate a clean pose stream including timestamps. b. Construct an event-marking driven engine to trigger feature update operations based on user interaction trigger areas: c employs differential coding technology in the incremental rendering stage: The feature hash comparison algorithm identifies the regions that need to be updated, and only the feature subsets with a hash value difference greater than 15% are regenerated; A progressive generation strategy is used: the first frame generates a low-resolution base mesh, and subsequent frames are gradually improved to the target resolution through residual connections; When the distributed synchronization engine performs multi-terminal push: Attach a device fingerprint identifier to each content frame, including screen color temperature calibration value and rendering pipeline latency compensation coefficient; A layered synchronization protocol is adopted: the base layer transmits geometric topology data, and the enhanced laminar transmission transmits lighting and shadow effects data; When version differences between terminals are detected, a content merging algorithm based on the minimum spanning tree is triggered.

8. The AI- and image detection-based art design virtual reality content synchronous display system according to claim 7, characterized in that, The execution logic of the event marker-driven engine is as follows: Analyze the boundary coordinates of the haptic feedback trigger area in the interactive response event markers; When the operation trajectory in the cleanup pose flow intersects with the trigger region: Calculate the feature set update amount ΔF based on the pressure-displacement curve; Generate update instruction packages with priority tags, the priority of which is determined by the heat value Hv of the interactive area.

Images