Visual intelligent interactive design system

Abstract

The invention relates to the field of intelligent interactive design, and particularly discloses a visual intelligent interactive design system which comprises a three-dimensional structure design module, an intelligent color matching module, a visual rendering module and an interactive feedback analysis module. The three-dimensional structure design module can process imported or custom a three-dimensional model, detect and repair topology errors and construct a three-dimensional scene basic framework; the intelligent color matching module generates a color matching scheme based on the instruction, a rule base or user preference and displays a comparison view; the visual rendering module renders a three-dimensional scene in real time, collects rendering pipeline load parameters and quality indexes, and analyzes balance points to adjust rendering parameters; and the interaction feedback analysis module records interaction data, marks over-limit nodes by comparing a standard and analyzes a root cause. According to the system, efficient three-dimensional scene construction, intelligent color matching, rendering optimization and interaction experience improvement are realized through multi-module cooperation, and the system is suitable for multi-field scenes needing visual intelligent interaction design.

Description

Technical Field

[0001] The present invention relates to the field of intelligent interaction design, and relates to a visual intelligent interaction design system. Background Art

[0002] With the rapid development of information technology, interaction design has been widely penetrated into multiple fields such as virtual reality, augmented reality, game development, industrial design, and architectural visualization. However, there are still some deficiencies in the existing interaction design systems in actual use, which restrict the improvement of user experience and the breakthrough of design efficiency.

[0003] On the one hand, the existing interaction design generally adopts fixed parameter configuration during rendering, and cannot optimize rendering parameters in real time according to the user's requirements for rendering quality and dynamically allocate resources. When processing high-complexity scenes, the video memory occupancy rate is too high, which is likely to cause program lag. Although reducing the rendering parameters can improve the fluency, it will lead to quality problems such as blurred image edges and incomplete color gamut coverage.

[0004] On the other hand, the interaction delay feedback mechanism of the existing interaction design is weak, and it cannot accurately identify over-limit interaction nodes and locate the root cause of over-limit interaction nodes, relying only on manual debugging, which is not conducive to targeted optimization. Summary of the Invention

[0005] In view of the above problems, the present invention proposes a visual intelligent interaction design system to realize the functions of intelligent interaction design.

[0006] The technical solution adopted by the present invention to solve its technical problems is: the present invention provides a visual intelligent interaction design system, including: a three-dimensional structure design module: identifying a three-dimensional model file imported by the user or a custom three-dimensional model and automatically detecting and repairing model topology errors, and generating a three-dimensional scene basic framework including spatial coordinate relationships.

[0007] An intelligent color matching module: generating a color matching scheme based on user input instructions, a preset color matching rule library or user preferences, and displaying a dynamic color matching scheme comparison view through an interface interaction component.

[0008] A visual rendering module: performing real-time rendering on the three-dimensional scene, synchronously collecting rendering pipeline load parameters and image quality evaluation indicators, analyzing the balance point between rendering efficiency and quality, and adaptively adjusting rendering parameters, where the rendering pipeline load parameters include video memory occupancy rate, shader execution cycle, and frame synchronization delay, and the image quality evaluation indicators include color gamut coverage rate and edge sharpness value.

[0009] An interaction feedback analysis module: recording interaction data of click response duration and animation transition delay in user operations, comparing the data with a preset fluency standard and marking over-limit interaction nodes, and performing root cause analysis of interaction delay according to the marking results.

[0010] Compared with the prior art, the visual intelligent interaction design system of the present invention has the following beneficial effects: 1. The present invention generates color matching schemes based on user input instructions, a preset color matching rule library, or user preferences. The coordination and interaction of the three modes enable the color matching link to cover the full-scenario color matching requirements from standardized rules to personalized needs, while improving the accuracy of scheme generation and user satisfaction.

[0011] 2. The present invention analyzes the balance point between rendering efficiency and quality by real-time monitoring of the rendering pipeline load parameters and image quality evaluation indicators, and adaptively adjusts the rendering parameters, and can flexibly adapt priorities according to different application scenarios, maximizing the use of existing resources to provide the best experience.

[0012] 3. The present invention realizes the rapid positioning and feedback of the root cause of latency by collecting the interaction data of user operations and combining the mapping relationship between interaction nodes and scene components, realizes targeted optimization, and avoids traditional trial-and-error modifications. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0014] Figure 1 It is the system module connection diagram of the present invention.

[0015] Figure 2 It is the system structure diagram of the present invention.

[0016] Figure 3 It is the working flow chart of the import model processing unit of the present invention.

[0017] Figure 4 It is the working flow chart of the custom model processing unit of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0019] Please refer to Figure 1 and Figure 2As shown in the figure, the present invention provides a visual intelligent interaction design system, including a three-dimensional structure design module, an intelligent color matching module, a visual rendering module, and an interaction feedback analysis module.

[0020] The intelligent color matching module is respectively connected to the three-dimensional structure design module and the visual rendering module, and the interaction feedback analysis module is connected to the visual rendering module.

[0021] The three-dimensional structure design module recognizes the three-dimensional model file imported by the user or the custom three-dimensional model and automatically detects and repairs the topological errors of the model, generating a basic framework of the three-dimensional scene containing spatial coordinate relationships.

[0022] Furthermore, the three-dimensional structure design module includes an imported model processing unit and a custom model processing unit. Refer to Figure 3 As shown in the figure, the specific working process of the imported model processing unit is as follows: A1. Model import and recognition: After the user selects and uploads a custom three-dimensional model file through the operation interface, the model data is parsed, and the key elements of the model are extracted to clarify the basic structure and characteristics of the model.

[0023] A2. Topological error detection: After the model recognition is completed, the topological error detection mechanism is automatically started, and the areas or elements with topological errors are marked through geometric calculation and data comparison algorithms.

[0024] A3. Topological error repair: Based on the detected topological errors, a preset repair algorithm is used for automatic repair.

[0025] A4. Spatial coordinate system establishment: According to the characteristics of the model and the user's settings, the parameters of the coordinate origin, axis directions, and unit length are determined, and a unified spatial coordinate system is established.

[0026] A5. Construction of the basic framework of the three-dimensional scene: After the above steps are completed, a basic framework of the three-dimensional scene containing spatial coordinate relationships is constructed based on the repaired three-dimensional model and combined with the established spatial coordinate system.

[0027] It should be noted that in three-dimensional model design, topological errors refer to the logical defects in the geometric connection relationships on the model surface or structure. Such errors will cause abnormalities in subsequent applications such as model rendering, animation, physical simulation, 3D printing, or engineering analysis.

[0028] As a preferred solution, in the topological error repair, for some simple errors, such as tiny gaps or overlapping patches, the system may repair them by adjusting the vertex positions, merging adjacent vertices or patches, etc.; for more complex non-manifold geometry problems, specific reconstruction algorithms may be used to reconstruct the relevant areas to ensure that the topological structure of the model meets the specifications and provides a good foundation for subsequent processing and rendering.

[0029] As a preferred solution, the specific process of constructing the basic framework of the three-dimensional scene is as follows: Integrate each part of the model according to its position and orientation in the spatial coordinate system to form a preliminary three-dimensional scene framework; At the same time, add necessary metadata and attribute information to the scene framework, such as the boundary range of the scene, lighting conditions, initial position and viewing angle of the camera, etc. These information will provide basic support for subsequent visual rendering, interactive operations, and the function implementation of other modules.

[0030] Further, referring to Figure 4 As shown, the specific working process of the custom model processing unit is as follows: B1. User modeling behavior capture: Parse the user's multi-modal input obtained from the dual channels of parametric modeling and interactive modeling to obtain the sketch contour drawn by the user and identify the basic geometric features of the sketch contour.

[0031] B2. Modeling processing: Perform corresponding parametric modeling processing and interactive modeling processing according to the type of multi-modal input.

[0032] B3. Topological structure pre-verification: Run incremental error detection in real time during the modeling process and provide dynamic repair solutions for the detected problems.

[0033] B4. Spatial relationship construction: Automatically generate a local coordinate system based on the initial modeling plane, establish an assembly relationship diagram through user-defined spatial constraints, implement a constraint propagation algorithm to solve the optimal solution of the global coordinate system and generate a hierarchical structure, and further automatically assign mass attributes according to the basic geometric features and optimize the material distribution of the feature structure.

[0034] B5. Dynamic construction of the scene framework: Sequentially execute operations such as real-time spatial indexing, lighting pre-adaptation, and multi-resolution management to dynamically construct the three-dimensional scene framework.

[0035] It should be noted that the dual recognition and topological repair capabilities of the three-dimensional structure design module can improve compatibility and flexibility, reduce manual intervention, and improve development efficiency.

[0036] The intelligent color matching module generates a color matching scheme based on user input instructions, a preset color matching rule library, or user preferences, and displays a dynamic color matching scheme comparison view through the interface interaction component.

[0037] Further, the intelligent color matching module includes a user-defined color matching unit, an automatic color matching unit, and a user preference color matching unit. The specific working process of the user-defined color matching unit is as follows: Parse the user's natural language instructions or parametric input to extract keywords, and convert them into computable color parameters in combination with the set color theory knowledge base.

[0038] Automatically generate color combination based on the parsing result, including single color, gradient color, and multi-color matching modes, and optimize the color scheme by combining with preset industry templates.

[0039] As a preferred solution, the keywords are such as "cool tone", "high saturation", "analogous color matching", etc., the color theory knowledge base is such as RGB / HSB color space, color wheel rules, etc., and the color parameters are such as hue range, brightness threshold, etc. In a specific embodiment, when the user inputs "generate a gradient scheme for blue-violet color system", the system parses it into a linear gradient with a hue of 240°-270° and a brightness of 40%-70%.

[0040] It should be noted that the core logic of generating a color scheme based on the user input instruction is to parse the user's natural language instruction or parametric input and map it to the color model to generate a scheme; the core logic of generating a color scheme based on the preset color matching rule library is to call the color combination model of the built-in rule library and quickly match the scheme according to the preset logic; the core logic of generating a color scheme based on the user preference is to train a personalized model through the user's historical operation data and dynamically generate a scheme that meets the preference.

[0041] It should be noted that the present invention generates color schemes based on the user input instruction, the preset color matching rule library, or the user preference respectively. The coordination and interaction of the three modes enable the color matching link to cover the full-scenario color matching requirements from standardized rules to personalized needs, and at the same time improve the accuracy of scheme generation and user satisfaction.

[0042] Furthermore, the specific working process of the automatic color matching unit is as follows: obtain the type of the three-dimensional scene and extract the key element features in the three-dimensional scene, including material attributes, lighting conditions, and the visually focused area marked by the user.

[0043] According to the key element features of the three-dimensional scene, compare its similarity with the historical case library under the same scene type in the preset color matching rule library, screen candidate color schemes from the preset color matching rule library according to the matching degree, and generate new schemes through genetic algorithm iteration. Further, display the schemes sorted by visual coordination degree in the user interface, and support the user to manually adjust the color values and preview the effects in real time.

[0044] Furthermore, the specific working process of the user preference color matching unit is as follows: obtain the user preference data according to the color matching scheme parameters, interaction behaviors, and scoring feedback selected by the user in the past, and construct user portrait labels.

[0045] Based on the collaborative filtering algorithm or deep learning model, analyze the correlation between the user preference data and the color parameters, generate a color scheme, and adjust the color scheme by updating the user preference data in real time.

[0046] As a preferred solution, the color scheme parameters include common primary color RGB values, preferred color tone tendencies, etc., the interaction behaviors include the dwell time on a certain color scheme, modification frequency, etc., the scoring feedback includes "like" "dislike" labels, etc., and the user portrait tags include "preference for warm colors", "commonly used high-contrast color schemes", etc.

[0047] In a specific embodiment, the user uses red as the primary color multiple times and prefers high saturation. The system automatically generates a high-contrast scheme with red as the keynote and the complementary color green.

[0048] It should be noted that the multi-source input advantage of the intelligent color matching module can combine personalization and standardization, reduce decision-making costs, and improve cross-domain applicability.

[0049] The visualization rendering module performs real-time rendering on the three-dimensional scene, synchronously collects the rendering pipeline load parameters and image quality evaluation indicators, analyzes the balance point between rendering efficiency and quality, and adaptively adjusts the rendering parameters. The rendering pipeline load parameters include video memory occupancy rate, shader execution cycle, and frame synchronization delay, and the image quality evaluation indicators include color gamut coverage rate and edge sharpness value.

[0050] Furthermore, the specific working process of the visualization rendering module includes: monitoring the total video memory occupied by the current rendering task and calculating its proportion of the total available video memory to obtain the video memory occupancy rate.

[0051] Insert a timestamp in the rendering pipeline to query the number of clock cycles from the start to the end of the shader and convert it to the actual elapsed time to obtain the shader execution cycle.

[0052] During the rendering process, calculate the difference between the time point when the frame is submitted to the GPU and the actual screen refresh time point to obtain the frame synchronization delay.

[0053] Intercept the rendering result as bitmap data and calculate the distribution ratio of pixels in the target color gamut through the GPU computing shader or the CPU-side image processing library, and calculate the proportion of the area that exceeds or does not cover the color gamut range to obtain the color gamut coverage rate.

[0054] On the GPU side, calculate the image gradient in real time through the post-processing channel to generate an edge intensity map, and statistically calculate the average gradient amplitude or peak signal-to-noise ratio in the edge area to obtain the edge sharpness value.

[0055] It should be noted that the reason for selecting the video memory occupancy rate, shader cycles, and frame synchronization latency as the rendering pipeline load parameters is that they are directly related to the GPU resource bottleneck, computing efficiency, and frame smoothness. The three respectively correspond to the storage, computing, and output links, forming a closed-loop performance monitoring system; the reason for selecting the color gamut coverage rate and edge sharpness value as the image quality evaluation indicators is that they quantify the two most sensitive experience dimensions of color authenticity and visual clarity, and the two together define visual fidelity.

[0056] Furthermore, the specific working process of the visual rendering module further includes: performing equal-gradient adjustment on the rendering pipeline load parameters in the set increasing order to obtain multiple sets of rendering pipeline load parameters and their corresponding image quality evaluation indicators.

[0057] Substitute multiple sets of rendering pipeline load parameters into the relationship model between the video memory occupancy rate, shader execution cycles, frame synchronization latency and rendering efficiency preset, to obtain multiple sets of data of the rendering efficiency. The relationship model includes the quantitative mapping relationship between the video memory occupancy rate, shader execution cycles, frame synchronization latency and rendering efficiency.

[0058] Similarly, substitute multiple sets of image quality evaluation indicators into the relationship model between the color gamut coverage rate and edge sharpness value and the rendering quality preset, to obtain multiple sets of data of the rendering quality.

[0059] Based on the multiple sets of data of the rendering efficiency and the rendering quality, analyze the correlation between the rendering efficiency and the rendering quality. Substitute the required value of the rendering quality into this correlation to obtain the corresponding rendering efficiency and record it as the balance point of the rendering efficiency and the quality. Based on this balance point, calculate the corresponding rendering pipeline load parameters and then adjust the rendering pipeline load parameters.

[0060] It should be noted that when the image quality is improved, the video memory occupancy rate will increase significantly, and the shader execution cycles will also grow non-linearly. At the same time, high-quality rendering leads to an increase in the single-frame rendering time, which in turn affects the synchronization latency, making the frame synchronization latency increase.

[0061] As a preferred solution, the adjustment of the rendering pipeline load parameters will not exceed its corresponding working range.

[0062] It should be noted that the core value of adaptively adjusting the rendering parameters based on the balance point of efficiency and quality lies in avoiding "performance overkill" or "insufficient picture quality" caused by fixed parameters, being able to flexibly adapt the priority according to different application scenarios, covering from low-end to high-end hardware, and maximizing the use of existing resources to provide the best experience; through the data feedback closed-loop, upgrading the traditional "static rendering pipeline" to an "adaptive rendering ecosystem", and finally achieving the two-way optimal solution of efficiency and quality.

[0063] It should be noted that the present invention monitors the load parameters of the rendering pipeline and the image quality evaluation indicators in real time, analyzes the balance point between rendering efficiency and quality, and adaptively adjusts the rendering parameters, which is conducive to dynamic resource optimization, avoiding performance bottlenecks, and quantifying balance decisions. It can flexibly adapt the priority according to different application scenarios and maximize the use of existing resources to provide the best experience.

[0064] The interaction feedback analysis module records the interaction data of the click response duration and the animation transition delay in the user operation, compares the data with the preset fluency standard, marks the over-limit interaction nodes, and analyzes the root cause of the interaction delay according to the marking result.

[0065] Further, the specific working process of the interaction feedback analysis module includes: recording the timestamp when the user click occurs through the user click event monitoring software on the front-end page, then listening to the moment when the response action triggered by the click event starts to execute and recording the timestamp at this time, and obtaining the click response duration in the user operation according to the difference between the two timestamps.

[0066] Recording the timestamps of the start of the animation and the key transition points through the animation event monitoring software, and obtaining the animation transition delay in the user operation by calculating the time difference between adjacent key frames.

[0067] It should be noted that the click response duration and the animation transition delay are selected as the operation data of user interaction. Both are core interaction quality indicators, and exceeding the human perception threshold will significantly reduce the satisfaction.

[0068] Further, the specific working process of the interaction feedback analysis module also includes: extracting the fluency standard stored in the database, obtaining the thresholds of the click response duration and the animation transition delay, comparing the interaction data of the user operation collected with the above thresholds, and if the click response duration or the animation transition delay of a certain interaction node exceeds its corresponding threshold, marking the interaction node as an over-limit interaction node.

[0069] Extracting the mapping relationship between the interaction node - scene component stored in the database and the set of problems existing in each scene component during operation, matching the scene component associated with the over-limit interaction node and the set of problems existing in its scene component during operation, and giving feedback.

[0070] In a specific embodiment, the threshold of the click response duration is set to 200 milliseconds, and the threshold of the animation transition delay is set to 100 milliseconds.

[0071] As a preferred solution, each interaction node is associated with its corresponding scene component during the system design phase. This mapping relationship can be recorded through code comments, configuration files, or data structures.

[0072] As a preferred solution, by combining the system's performance monitoring data and code logs, potential problems of each scenario component during operation can be obtained.

[0073] In a specific embodiment, if the click response of a certain button times out, the corresponding scenario component is a complex form submission component, which may be caused by reasons such as overly complex form validation logic, network request latency, or memory leak.

[0074] As a preferred solution, further simulate the user's operations according to the recorded operation steps and data of the over-limit interaction nodes to reproduce the lag problem. Analyze the execution process of the scenario component step by step through the debugging tool to find out the specific code or logic that causes the delay.

[0075] It should be noted that the present invention combines the interaction data collected from the user's operations with the mapping relationship between the interaction nodes and the scenario components to quickly locate and feedback the root cause of the delay, achieve targeted optimization, and avoid traditional trial-and-error modifications.

[0076] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0077] Those of ordinary skill in the art can realize that the modules and algorithm steps of the examples described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0078] In addition, in each embodiment of the present application, the various functional modules can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0079] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in this application, and all should be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

[0080] Finally, the above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included in the protection scope of the present invention.

Claims

1. A visual intelligent interaction design system, characterized in that, Including: 3D Structure Design Module: Identify the 3D model file imported by the user or the custom 3D model, automatically detect and repair the topological errors of the model, and generate the basic framework of the 3D scene containing spatial coordinate relationships; Intelligent Color Matching Module: Generate color matching schemes based on user input instructions, preset color matching rule libraries or user preferences, and display a dynamic color matching scheme comparison view through the interface interaction component; Visualization Rendering Module: Render the 3D scene in real time, synchronously collect the load parameters of the rendering pipeline and the image quality evaluation indicators, analyze the balance point between rendering efficiency and quality, and adaptively adjust the rendering parameters. The load parameters of the rendering pipeline include video memory occupancy rate, shader execution cycle, and frame synchronization delay, and the image quality evaluation indicators include color gamut coverage rate and edge sharpness value; Interactive Feedback Analysis Module: Record the interaction data of click response duration and animation transition delay in the user operation, compare the data with the preset smoothness standard and mark the over-limit interaction nodes, and perform root cause analysis of interaction delay according to the marking results.

2. The visual intelligent interaction design system according to claim 1, characterized in that: The 3D Structure Design Module includes an imported model processing unit and a custom model processing unit. The specific working process of the imported model processing unit is as follows: A1. Model Import and Recognition: After the user selects and uploads a custom 3D model file through the operation interface, the model data is parsed, and the key elements of the model are extracted to clarify the basic structure and characteristics of the model; A2. Topological Error Detection: After the model recognition is completed, the topological error detection mechanism is automatically started, and the areas or elements with topological errors are marked through geometric calculation and data comparison algorithms; A3. Topological Error Repair: Automatically repair based on the detected topological errors using the preset repair algorithm; A4. Spatial Coordinate System Establishment: Determine the parameters of the coordinate origin, coordinate axis direction, and unit length according to the characteristics of the model and the user's settings, and establish a unified spatial coordinate system; A5. Construct the Basic Framework of the 3D Scene: After completing the above steps, construct the basic framework of the 3D scene containing spatial coordinate relationships based on the repaired 3D model and combined with the established spatial coordinate system.

3. The visual intelligent interaction design system according to claim 2, wherein: The specific working process of the custom model processing unit is as follows: B1. User Modeling Behavior Capture: Parse the user's multi-modal input obtained through the dual channels of parametric modeling and interactive modeling to obtain the sketch contour drawn by the user and identify the basic geometric features of the sketch contour; B2. Modeling Processing: Perform corresponding parametric modeling processing and interactive modeling processing according to the type of multi-modal input; B3. Topological Structure Pre-Verification: Run incremental error detection in real time during the modeling process, and provide dynamic repair solutions for the detected problems; B4. Spatial Relationship Construction: Automatically generate a local coordinate system based on the initial modeling plane, establish an assembly relationship diagram through the spatial constraints defined by the user, implement the constraint propagation algorithm to solve the optimal solution of the global coordinate system and generate a hierarchical structure, and further automatically assign mass attributes according to the basic geometric features and optimize the material distribution of the feature structure; B5. Dynamic Construction of the Scene Framework: Sequentially perform operations such as real-time spatial indexing, light pre-adaptation, and multi-resolution management to dynamically construct the 3D scene framework.

4. A visual intelligent interaction design system according to claim 1, characterized in that: The intelligent color matching module includes a user-defined color matching unit, an automatic color matching unit, and a user preference color matching unit. The specific working process of the user-defined color matching unit is as follows: Parse the natural language instructions or parametric inputs of the user to extract keywords, and convert them into computable color parameters in combination with the set color theory knowledge base; Automatically generate color matching combinations based on the parsing results, including single-color, gradient color, and multi-color matching modes, and optimize the color matching scheme in combination with the preset industry templates.

5. A visual intelligent interaction design system according to claim 4, characterized in that: The specific working process of the automatic color matching unit is as follows: Obtain the type of the three-dimensional scene and extract the key element features in the three-dimensional scene, including material properties, lighting conditions, and the visually focused area marked by the user; Compare the key element features of the three-dimensional scene with the historical case library of the same scene type in the preset color matching rule library for similarity, screen candidate color matching schemes from the preset color matching rule library according to the matching degree, and iteratively generate new schemes through a genetic algorithm. Further, display the schemes sorted by visual coordination degree in the user interface, and support the user to manually adjust the color values and preview the effects in real time.

6. A visual intelligent interaction design system according to claim 4, characterized in that: The specific working process of the user preference color matching unit is as follows: Obtain user preference data based on the color matching scheme parameters, interaction behaviors, and rating feedback selected by the user in the past, and construct user portrait tags; Based on a collaborative filtering algorithm or a deep learning model, analyze the correlation between user preference data and color parameters, generate a color matching scheme, and adjust the color matching scheme in real time by updating the user preference data.

7. A visual intelligent interaction design system according to claim 1, characterized in that: The specific working process of the visualization rendering module includes: Monitor the total video memory occupied by the current rendering task and calculate its proportion of the total available video memory to obtain the video memory occupancy rate; Insert a timestamp query record in the rendering pipeline to calculate the number of clock cycles of the shader from start to end and convert it into the actual elapsed time to obtain the shader execution cycle; During the rendering process, calculate the difference between the time point when the frame is submitted to the GPU and the actual screen refresh time point to obtain the frame synchronization delay; Intercept the rendering result as bitmap data and calculate the distribution ratio of pixels in the target color gamut through a GPU computing shader or a CPU-side image processing library, and calculate the proportion of the area exceeding or not covering the color gamut range to obtain the color gamut coverage rate; On the GPU side, calculate the image gradient in real time through a post-processing channel to generate an edge intensity map, and statistically calculate the average gradient amplitude or peak signal-to-noise ratio in the edge area to obtain the edge sharpness value.

8. A visual intelligent interaction design system according to claim 1, characterized in that: The specific working process of the visualization rendering module also includes: Adjust the rendering pipeline load parameters in equal gradients according to the set increasing order to obtain multiple sets of rendering pipeline load parameters and their corresponding image quality evaluation indicators; Substitute multiple sets of rendering pipeline load parameters into the relationship model between the video memory occupancy rate, shader execution cycle, frame synchronization delay, and rendering efficiency to obtain multiple sets of data on rendering efficiency. The relationship model includes the quantitative mapping relationship between the video memory occupancy rate, shader execution cycle, frame synchronization delay, and rendering efficiency; Similarly, substitute multiple sets of image quality evaluation indicators into the relationship model between the color gamut coverage rate and edge sharpness value and rendering quality to obtain multiple sets of data on rendering quality; Analyze the correlation between rendering efficiency and rendering quality based on multiple sets of data of rendering efficiency and rendering quality, substitute the required value of rendering quality into this correlation to obtain the corresponding rendering efficiency and record it as the balance point of rendering efficiency and quality, and then infer the corresponding rendering pipeline load parameters based on this balance point and adjust the rendering pipeline load parameters accordingly.

9. A visual intelligent interaction design system according to claim 1, characterized in that: The specific working process of the interaction feedback analysis module includes: Record the timestamp when the user click occurs through the software that listens to the user click event in the front-end page, then listen to the moment when the response action triggered by this click event starts to execute and record the timestamp at this time, and obtain the click response duration in the user operation according to the difference between the two timestamps; Record the timestamps of the start and key transition points of the animation through the software that listens to the animation event, and obtain the animation transition delay in the user operation by calculating the time difference between adjacent key frames.

10. A visual intelligent interaction design system according to claim 9, characterized in that: The specific working process of the interaction feedback analysis module further includes: Extract the fluency standard stored in the database, obtain the thresholds of the click response duration and the animation transition delay, compare the interaction data of the user operation collected with the above thresholds, and if the click response duration or the animation transition delay of a certain interaction node exceeds its corresponding threshold, mark this interaction node as an over-limit interaction node; Extract the mapping relationship between the interaction node - scene component stored in the database and the set of problems existing in each scene component during operation, match the scene component associated with the over-limit interaction node and the set of problems existing in its scene component during operation, and give feedback.

Images