A method and system for generating public art design schemes based on knowledge graphs

Abstract

This application provides a method and system for generating public art design schemes based on knowledge graphs, relating to the field of public art design technology. This application constructs a knowledge graph containing entities with design intent and semantic relationships. It acquires the cultural and emotional text of the target public space, identifies emotional tendencies, and matches them with the design intent entities in the knowledge graph. Combining the physical layout, ambient lighting, and acoustic characteristics of the target public space, it performs environmental adaptation processing on the initial set of design components to obtain a target set of design components. Based on a pre-set plot development logic, the target design components are organized and ordered to form a narrative structure sequence with defined artistic expression temporal and state transition relationships. Based on this sequence, it deconstructs and parametrically transforms to generate virtual scene data related to the real space. Through augmented reality fusion rendering, it generates an art experience scheme, enabling the generation of augmented reality art experience schemes that are deeply adapted to the spatial culture and physical environment.

Description

Technical Field

[0001] This application relates to the field of public art design technology, and in particular to a method and system for generating public art design schemes based on knowledge graphs. Background Technology

[0002] Public art design schemes are an important carrier of cultural expression in public spaces, which can enhance the aesthetic value and humanistic atmosphere of spaces. They have broad application prospects in urban squares, parks, transportation hubs and other scenarios, and their intelligent generation methods have attracted much attention from the industry.

[0003] Currently, existing public art design technologies based on knowledge graphs mostly integrate design elements by constructing domain-specific knowledge graphs to provide data support for scheme generation. Some methods have attempted to initially associate and match knowledge graphs with design requirements. However, these technologies often focus on the simple listing and retrieval of design elements, lacking in-depth consideration of adaptation to specific scenarios.

[0004] However, existing methods struggle to fully integrate the physical environmental characteristics and cultural and emotional needs of the target public space, resulting in design schemes that are insufficiently adapted to the space, have a poor user experience, and fail to achieve an organic unity between artistic expression and scene atmosphere. Therefore, existing technologies suffer from the technical problem of insufficient adaptability between public art design schemes and spatial scenes and cultural emotions. Summary of the Invention

[0005] The purpose of this application is to provide a method and system for generating public art design schemes based on knowledge graphs, so as to solve the problem of insufficient adaptability of schemes to spatial scenes and cultural emotions in existing technologies.

[0006] To address the aforementioned technical problems, firstly, this application provides a method for generating public art design schemes based on knowledge graphs, comprising:

[0007] A knowledge graph is constructed based on design element information in the field of public art. The knowledge graph includes entities that express design intentions and semantic relationships that connect the entities.

[0008] The process involves acquiring text content that reflects the cultural sentiment of the target public space, identifying sentiment information from the text content, associating and matching the sentiment information with entities in the knowledge graph that express design intent, and determining an initial set of design components from the knowledge graph based on the matching results.

[0009] Based on the physical layout, ambient lighting, and acoustic characteristics of the target public space, environmental adaptability processing is performed on the elements in the initial design component set to determine the target design component set.

[0010] According to the preset plot development logic, the components in the target design component set are organized and sorted to form a narrative structure sequence, which specifies the temporal and state transition relationships of different artistic expressions in the experience process;

[0011] Based on the artistic expressions in the narrative structure sequence, virtual scene data associated with real spatial locations is generated through deconstruction and parametric transformation. Augmented reality fusion rendering is then performed on the virtual scene data to generate an artistic experience scheme.

[0012] Optionally, the elements in the target design component set are organized and sorted according to a preset plot development logic to form a narrative structure sequence, including:

[0013] Define the plot development logic, define the corresponding theme and emotional tone for each stage of the plot development logic, and classify the components in the target design component set according to their artistic expression.

[0014] Based on the theme and emotional tone, matching components are selected from the categorized components, and the display order and duration within the corresponding plot stage are determined according to the attributes of each component, so as to generate the component sequence corresponding to each stage.

[0015] The component sequences corresponding to each stage are connected in chronological order according to the plot development logic, and state transition rules are configured at the junctions of adjacent sequences to form a narrative structure sequence.

[0016] Optionally, the component sequences corresponding to each stage are connected according to the chronological order of the plot development logic, and state transition rules are configured at the junctions of adjacent sequences to form a narrative structure sequence, including:

[0017] Extract the start and end time markers of the component sequences corresponding to each stage, determine the order of each sequence according to the plot development logic, and arrange them sequentially on the timeline;

[0018] Analyze the attribute differences between the last component and the first component in adjacent sequences, and calculate the intermediate attribute value sequence required for a smooth transition based on the attribute differences;

[0019] Insert a state transition instruction at the junction of adjacent sequences, wherein the state transition instruction includes the intermediate attribute value sequence and the corresponding transition time length;

[0020] The chronologically ordered sequence is combined with the state transition instructions to generate a narrative structure sequence.

[0021] Optionally, based on the artistic expressions in the narrative structure sequence, virtual scene data associated with real spatial locations is generated through deconstruction and parametric transformation. Augmented reality fusion rendering processing is then performed on the virtual scene data to generate an artistic experience scheme, including:

[0022] The artistic expressions in the narrative structure sequence are decomposed into virtual elements, which include graphics, color values, motion trajectories and time markers, and the virtual elements are converted into parametric models.

[0023] The parameterized model is associated with the real spatial location, and the position of the virtual element is mapped to a specific point in the corresponding physical space through spatial anchoring technology, thereby generating virtual scene data with spatial association.

[0024] Based on augmented reality rendering rules, the transparency, shadows, and lighting effects of the virtual scene data are adjusted to blend with the real environment;

[0025] Based on the timeline of the narrative structure sequence, the rendering output of the virtual scene data is triggered sequentially to generate an art experience scheme, which includes virtual content playback instructions with time stamps.

[0026] Optionally, the parameterized model is associated with a real spatial location. Using spatial anchoring technology, the position of the virtual element is mapped to a specific point in the corresponding physical space, generating a group of virtual elements with spatial association, including:

[0027] Based on the position parameters in the parameterized model and the coordinate data of the real space, multiple physical points are selected in the real space as spatial anchor points, and the physical coordinates and corresponding virtual coordinates are marked for the spatial anchor points.

[0028] Based on the physical coordinates and virtual coordinates, coordinate transformation parameters are calculated, and a transformation relationship from the virtual coordinate system to the real coordinate system is established. The transformation relationship includes a rotation matrix and a translation vector.

[0029] By applying the transformation relationship, the coordinates of the virtual element in the parameterized model are transformed to obtain the corresponding coordinates of the virtual element in the real space, and the position of the virtual element is adjusted to match a specific point in the physical space.

[0030] All the virtual elements with adjusted positions are grouped according to spatial proximity to generate virtual element groups with spatial relationships.

[0031] Optionally, text content reflecting the cultural sentiment of the target public space is acquired, sentiment information is identified from the text content, and the sentiment information is matched with entities expressing design intent in the knowledge graph. Based on the matching results, a set of design components is determined from the knowledge graph, including:

[0032] The text content of the target public space is obtained, words with cultural characteristics and emotional states are extracted from the text content to form a vocabulary group, and emotional values ​​are assigned to the words in the vocabulary group.

[0033] The words in the vocabulary group are scored on multiple dimensions, and all the scores are combined to generate sentiment information, which includes sentiment vectors.

[0034] In the knowledge graph, entities expressing design intent are labeled with sentiment attributes, the degree of matching between the sentiment vector and the sentiment attribute is calculated, and matching entities are selected based on the degree of matching.

[0035] The matching entities are used as candidate components. Based on the relationships between entities in the knowledge graph, closely related adjacent entities are added to the candidate components to form an initial design component set.

[0036] Optionally, based on the physical layout, ambient lighting, and acoustic characteristics of the target public space, environmental adaptability processing is performed on the elements in the initial design component set to determine the target design component set, including:

[0037] Acquire environmental information of the target public space, including physical layout, ambient lighting and acoustic characteristics;

[0038] Extract component parameters from the initial design component set, including spatial size requirements, light response characteristics, and sound interaction parameters;

[0039] The component parameters are verified to be compatible with the environmental information, mismatches between the components and the environment are identified and eliminated, and the visual performance parameters and acoustic output parameters of the components are adaptively adjusted.

[0040] The adjusted components are integrated to generate a set of target design components that are adapted to the environment.

[0041] Secondly, this application provides a public art design scheme generation system based on knowledge graphs, including:

[0042] A construction module is used to build a knowledge graph based on design element information in the field of public art. The knowledge graph contains entities that express design intentions and semantic relationships that connect the entities.

[0043] The matching module is used to acquire text content that reflects the cultural sentiment of the target public space, identify sentiment information from the text content, and associate and match the sentiment information with entities in the knowledge graph that express design intentions. Based on the matching results, the initial design component set is determined from the knowledge graph.

[0044] The processing module is used to perform environmental adaptability processing on the elements in the initial design component set based on the physical layout, ambient lighting and acoustic characteristics of the target public space, and to determine the target design component set.

[0045] The sorting module is used to organize and sort the components in the target design component set according to the preset plot development logic to form a narrative structure sequence, wherein the narrative structure sequence specifies the temporal and state transition relationships of different artistic expressions in the experience process;

[0046] The generation module is used to generate virtual scene data associated with real spatial locations based on the artistic expressions in the narrative structure sequence through deconstruction and parametric transformation, and to perform augmented reality fusion rendering processing on the virtual scene data to generate an art experience scheme.

[0047] Thirdly, this application provides an electronic device, comprising:

[0048] Memory, used to store computer programs;

[0049] A processor, configured to execute the computer program to implement the steps of the knowledge graph-based public art design scheme generation method as described in the first aspect above.

[0050] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the knowledge graph-based public art design scheme generation method described in the first aspect above.

[0051] The knowledge graph-based public art design scheme generation method provided in this application constructs a knowledge graph containing design intent entities and semantic relationships based on design elements in the public art field, providing structured and interconnected domain knowledge support for scheme generation. By acquiring the cultural and emotional text of the target public space and identifying emotional tendencies, and then associating and matching it with entities in the knowledge graph to determine the initial set of design components, the method achieves precise alignment between the design scheme and the cultural and emotional needs of the space. By combining the physical layout, lighting, and acoustic characteristics of the target space to perform environmental adaptability processing on the initial components to determine the target component set, the method improves the adaptability of the design components to the actual spatial environment. By sorting the target components according to a preset plot logic to form a narrative structure sequence, the method clarifies the temporal and state transition relationships of artistic expression forms, giving the scheme a narrative quality. By deconstructing and parametrically transforming the narrative structure sequence to generate virtual scene data related to the real space and performing augmented reality fusion rendering, the method generates a public art scheme that can be experienced immersively.

[0052] Furthermore, the plot development logic is established, and the themes and emotional tones of each stage are defined. Target design components are categorized according to their artistic expression forms. Matching components are selected based on the themes and emotional tones, and their display order and duration in the corresponding plot stages are determined. This generates a sequence of components for each stage. Then, the sequences are connected according to the plot sequence, and state transition rules are configured at the junctions of adjacent sequences to form a narrative structure sequence. Through meticulous plot logic setting and component sequence arrangement, the orderly connection of artistic expression forms and the coherent progression of emotional expression are achieved, ensuring the integrity and smoothness of the public art experience process. Attached Figure Description

[0053] To more clearly illustrate the technical solutions of the embodiments of this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart illustrating a public art design scheme generation method based on knowledge graphs, provided for an embodiment of this application;

[0055] Figure 2 A flowchart illustrating the specific implementation of a knowledge graph-based public art design scheme generation method provided in this application embodiment;

[0056] Figure 3 This is a schematic diagram of the structure of a public art design scheme generation system based on knowledge graphs, provided in an embodiment of this application. Detailed Implementation

[0057] Existing public art design technologies based on knowledge graphs often focus on the integration and initial matching of design elements, neglecting in-depth consideration of the physical environment and cultural emotions of the target space. This technological limitation results in design solutions that can only achieve a simple accumulation of elements, making it difficult to match with the physical layout, lighting, and acoustic characteristics of the space. At the same time, it fails to accurately convey the cultural and emotional connotations of the space, ultimately leading to problems such as insufficient adaptability of the solution to the scene and a poor artistic experience.

[0058] To address the aforementioned issues, this application proposes a knowledge graph-based method for generating public art design schemes. Its core lies in constructing a knowledge graph containing entities with design intent and semantic relationships. First, spatial cultural and emotional needs are matched with entities in the knowledge graph to determine initial design components. Then, the components are adapted to the spatial physical characteristics to achieve environmental adaptation. Subsequently, a narrative structure sequence is formed according to plot logic. Finally, an art experience scheme is generated through parametric transformation and augmented reality rendering. This method, through the dual constraints of cultural and emotional matching and environmental adaptation, as well as the dual optimizations of narrative arrangement and augmented reality presentation, fundamentally solves the problem of insufficient adaptability of schemes to spatial scenes and cultural emotions in existing technologies, achieving a deep integration of public art design and space.

[0059] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0060] The core of this application is to provide a method for generating public art design schemes based on knowledge graphs. A flowchart of one specific implementation method is shown below. Figure 1 As shown, the method includes:

[0061] S101. Constructing a knowledge graph based on design element information in the field of public art;

[0062] In this step, a knowledge graph is a structured form of knowledge storage and representation, containing entities that express design intent and semantic relationships connecting these entities. Semantic relationships are used to describe the logical connections between different entities, such as a certain type of cultural symbol corresponding to a specific art style, or a certain material being suitable for a specific form of expression.

[0063] In one specific implementation, the core design elements of the public art field are first comprehensively collected through methods such as literature review, industry case analysis, and designer experience summaries. The specific attributes of each element are clarified, such as artistic style, color tendency, material properties, and durability. Next, the inherent relationships between elements are analyzed to determine the types of relationships, such as causal relationships, adaptive relationships, and subordinate relationships. Finally, these elements are transformed into entities in a knowledge graph, and the logical relationships between elements are converted into semantic relationships between entities. A knowledge graph construction tool is used to complete the structured organization, forming a knowledge graph that can support the generation of design solutions.

[0064] For example, in the context of a red-themed urban plaza, the core elements are first collected by reviewing literature on red-themed public art design, analyzing domestic cases of red plaza art layouts, and summarizing designers' practical experience in red-themed projects. These elements include forms of expression such as sculptures and reliefs, artistic styles such as solemnity and loyalty, materials such as granite and red LED light sources, and cultural symbols such as five-pointed stars and red flags.

[0065] Next, the specific attributes of each element are clarified. The sculpture height and relief area are categorized as spatial size requirements; the red LED light source's compatibility with the red spectrum represents its light response characteristics; and the outdoor material's weather resistance of no less than 15 years is a weather resistance indicator. Then, the types of relationships between the elements are analyzed. The sculpture and the solemn and dignified style are a compatibility relationship; the granite's hard texture and symbolism of immortality form a causal relationship with the outdoor commemorative sculpture; the red LED light source and the red-themed projection are a subordinate relationship; and the interactive commemorative installation and the slogan are a triggering relationship.

[0066] Next, these forms of expression, artistic styles, materials, and cultural symbols are transformed into graph entities, and various relationships are converted into semantic relationships between entities. At the same time, relationship weights are assigned, such as a weight of 0.9 for the compatibility relationship between sculpture and a solemn and dignified style, and a weight of 0.85 for the support relationship between granite and sculpture. Finally, graph building tools such as Neo4j are used to complete the structured organization, forming a knowledge graph that supports the public art design of the city's red-themed square.

[0067] The graph construction tools used can be mainstream graph database tools such as Neo4j and Arango DB. These tools support the visual modeling and storage of entities and semantic relationships, and can quickly establish the association index between entities, making it convenient to efficiently call data when performing sentiment matching and component retrieval.

[0068] S102. Obtain text content reflecting the cultural sentiment of the target public space, identify sentiment information from the text content, and match the sentiment information with entities expressing design intent in the knowledge graph. Determine the initial design component set from the knowledge graph based on the matching results.

[0069] Among them, sentiment information is a comprehensive description of the cultural connotations and emotional tone conveyed by the text content, including sentiment vectors that can quantify emotional characteristics. The initial design component set is a set of basic elements that can support public art design, selected through sentiment matching, and includes core entities and related entities in the knowledge graph that are compatible with the spatial culture and emotion.

[0070] Optionally, step S102 may specifically include the following steps:

[0071] S1021. Obtain the text content of the target public space, extract words with cultural characteristics and emotional states from the text content, form a vocabulary group, and assign emotional values ​​to the words in the vocabulary group.

[0072] Specifically, the process begins by collecting textual materials such as local chronicles, promotional materials, and resident interview records from the target public space. From these texts, words reflecting regional cultural characteristics and emotional states are extracted. For example, using a city's red-themed square as a scenario, researchers first visit the red history propaganda boards, exhibition halls, and model monuments within the square to collect information on local battles recorded on the propaganda boards, the heroic deeds of martyrs explained in the exhibition halls, and the inscriptions on the monuments. They then collect information from interviews with the public, recording their expressions of respect for the martyrs and their descriptions of historical memories. Next, from these collected textual materials and interview records, core words reflecting the spatial cultural characteristics and emotional tone are selected and refined. Finally, words such as "red memory," "martyrs," "heroic and fearless," "historical heritage," "solemn reverence," and "loyalty and perseverance" are extracted to form a vocabulary group. Then, based on an emotional dictionary, an emotional value is assigned to each word: positive emotional words are assigned positive values, negative emotional words are assigned negative values, and neutral emotional words are assigned zero values.

[0073] The sentiment dictionary used can adopt mature dictionary databases such as CNKI's How Net sentiment dictionary and Boson NLP sentiment dictionary, or it can be expanded with custom vocabulary in combination with the characteristics of the public art field to ensure that the sentiment values ​​of cultural sentiment words, such as the sentiment values ​​of red memory and historical heritage, are more in line with the field scenario.

[0074] S1022. The words in the vocabulary group are scored in multiple dimensions, and sentiment information is generated by combining all the scoring results. The sentiment information includes a sentiment vector.

[0075] Among them, sentiment vector is a vector data formed by quantifying and integrating the multi-dimensional scoring results of words, which is used to accurately represent the sentiment characteristics of the text.

[0076] In one specific implementation, three dimensions are selected: emotional intensity, cultural fit, and scene suitability. The Likert five-point scoring method is used to score each word in the vocabulary group separately.

[0077] For example, historical depth scores 3 points in emotional intensity, 4 points in cultural fit, and 4 points in scene suitability; solemnity scores 4 points in emotional intensity, 3 points in cultural fit, and 4 points in scene suitability. Then, a weighted summation formula is used to calculate the overall score for each word. The formula is as follows:

[0078] ;

[0079] in, Indicates the first The overall score of each word; , , These are the weights for the three dimensions, and their values ​​are all... ; The scores are given to the words in three dimensions.

[0080] Taking historical heritage as an example, , The solemn calculation process is as follows: , The overall scores of all words are arranged in order to form an sentiment vector, for example […]. , , , ] .

[0081] In another specific implementation, dimensions such as cultural uniqueness can be added according to actual needs, and the weights of each dimension can be adjusted. This application does not limit this. The above example is only one example of this application. In practical applications, settings can be made according to needs, and this application does not limit this.

[0082] S1023. In the knowledge graph, mark the entities that express the design intent with emotional attributes, calculate the matching degree between the emotional vector and the emotional attribute, and select matching entities according to the matching degree;

[0083] Among them, the sentiment attribute is a sentiment feature identifier assigned to entities in the knowledge graph that express design intent. It corresponds to the dimension of the sentiment vector and is used to achieve sentiment-level association matching.

[0084] In this embodiment, each entity expressing design intent in the knowledge graph is first labeled with emotional attributes. For example, traditional sculpture entities are labeled with emotional attributes related to solemnity and elegance; modern installation art entities are labeled with emotional attributes related to vitality and innovation, and these emotional attributes are quantified into attribute vectors consistent with the emotional vector dimension.

[0085] Next, the cosine similarity algorithm is used to calculate the matching degree between the sentiment vector and the attribute vectors of each entity. The formula is as follows:

[0086] ;

[0087] in, Represents the sentiment vector. Represents an entity attribute vector. Represents the dot product of two vectors. The magnitude of the sentiment vector, Represents the modulus of the entity attribute vector.

[0088] The core of the cosine similarity algorithm is to measure the degree of similarity by calculating the cosine of the angle between two vectors, with a value ranging from -1 to 1. The closer the value is to 1, the higher the matching degree. This algorithm has mature applications in text sentiment matching and feature vector comparison scenarios, and can effectively quantify the degree of emotional compatibility. Finally, based on the calculated matching degree, matching entities that meet the requirements are selected.

[0089] In practical applications, assuming the generated sentiment vector The entity attribute vector of a sculpture in a knowledge graph Next, the degree of matching of the sculptures is calculated. The calculation process is as follows: ; ; ; The final matching degree of the sculpture was 0.999. With a matching threshold of 0.8, the sculpture was selected as one of the matching entities if its matching degree exceeded the preset threshold. All selected matching entities were used as core design components. Subsequently, based on the relationships in the knowledge graph, related elements such as stone materials and relief techniques were added to form the initial design component set. The above example is merely one illustration of this application; in practical applications, settings can be customized according to requirements, and this application does not impose any limitations on this.

[0090] S1024. Using the matching entity as a candidate component, and based on the relationship between entities in the knowledge graph, add closely related adjacent entities to the candidate component to form an initial design component set.

[0091] In one specific implementation, the matching entities selected in S1023 are first identified as candidate components. Then, by analyzing the semantic relationships between entities in the knowledge graph and querying the association weights between entities, adjacent entities with weights higher than a set threshold are selected. These adjacent entities are key supporting elements for the implementation of candidate components. Finally, the candidate components are integrated with closely related adjacent entities to form an initial design component set with complete elements and logical connections.

[0092] For example, in a city's red-themed plaza scene, the previously selected sculptures and other matching entities are first identified as initial candidate components. These initial candidate components are core artistic expressions that align with the plaza's red theme. Then, relying on the knowledge graph of the city's red-themed plaza, the semantic relationships and weights corresponding to the sculpture entities are queried. Specifically, the support relationship between the sculpture and granite material has a weight of 0.85, and the compatibility relationship with the dignified styling technique has a weight of 0.9; both weights are higher than the set threshold of 0.7. Based on the weight selection results, granite material and dignified styling technique are determined to be closely related adjacent entities of the sculpture. Finally, the candidate components are integrated with the selected adjacent entities to form an initial design component set containing the sculpture, granite material, and dignified styling technique.

[0093] This application addresses the problems of inaccurate cultural and emotional expression and the disconnect between design elements and spatial connotations caused by the reliance on subjective experience in traditional public art design by quantifying the cultural and emotional characteristics of texts and accurately matching them with entities in a knowledge graph. Compared to the traditional approach of simply piling up elements, this step achieves a quantitative correlation between cultural emotions and design elements, ensuring that design components not only fit the cultural positioning of the target space but also possess logical coherence, significantly improving the cultural adaptability and scientific rigor of the design scheme.

[0094] S103. Based on the physical layout, ambient lighting and acoustic characteristics of the target public space, perform environmental adaptability processing on the elements in the initial design component set to determine the target design component set.

[0095] Environmental adaptation is the process of adjusting and optimizing the parameters of design components based on the actual environmental conditions of the target public space to eliminate mismatches. The target design component set is a collection of components that, after environmental adaptation adjustments, can effectively integrate with the spatial environment to achieve the desired artistic effect.

[0096] Optionally, step S103 may specifically include the following steps:

[0097] S1031. Obtain environmental information of the target public space, wherein the environmental information includes physical layout, ambient lighting and acoustic characteristics.

[0098] In one specific implementation, a laser ranging device scans the target public space to obtain physical layout information, including data such as space size, area division, and obstacle distribution; an illumination sensor collects ambient illumination information at different times, including data such as illumination intensity, illumination direction, and color temperature; and an acoustic detector collects acoustic characteristic information, including data such as ambient background noise decibel value and sound reflection coefficient.

[0099] For example, the core functional areas of the square were first identified, and then targeted data collection was carried out. In terms of physical layout, a laser rangefinder was used to scan the entire square, which was measured to be 90 meters long and 60 meters wide. The central area was the main venue for the commemorative activities, the north side was the sculpture placement area, the south side was the relief display wall, and the east and west sides were green landscape belts. At the same time, detailed data such as the height of the green plants and the size of the reserved base for the sculpture placement area were recorded.

[0100] Regarding ambient lighting, three time periods were selected: 8:00 AM, 12:00 PM, and 5:00 PM. For each time period, three sets of data were collected from three representative locations: the center of the square, the sculpture area, and the relief wall. The final average value was taken. The results are as follows: morning light intensity 3200 lux, color temperature 4800 K; noon light intensity 8500 lux, color temperature 5600 K; evening light intensity 2800 lux, color temperature 4200 K. It was also noted that the noon light mainly came from the due south direction and did not shine directly on the relief display wall.

[0101] In terms of acoustic characteristics, data were continuously collected for 5 minutes at 5 representative locations in the center of the square and next to the green plants at the four corners. The average background noise was measured to be 42 decibels. Since there were no high noise pollution sources around the square, the average sound reflection coefficient was 0.35. The reflection coefficient was slightly higher in the central area and lower in the green plant area.

[0102] S1032. Extract component parameters from the initial design component set;

[0103] The component parameters include space size requirements, light response characteristics, and sound interaction parameters. Space size requirements refer to the length, width, and height specifications of the component itself, as well as the space reserved for installation. Light response characteristics refer to the component's adaptability to ambient light and the parameters related to its presentation effect. Sound interaction parameters refer to the relevant technical indicators when the component has sound output or sensor interaction functions.

[0104] Specifically, key parameters for each component are extracted one by one from the initial design component set. Extracting component parameters relies on the core elements of the initial design component set, combined with public art design specifications, material technical standards, and component functional requirements. First, the core parameter dimensions of each component are identified. Then, by consulting design manuals, accessing pre-set component attribute data in the knowledge graph, and combining industry technical standards, the specific parameter values ​​for each dimension are determined one by one, ensuring that the parameters not only meet the component's functional requirements but also possess quantifiable and verifiable characteristics.

[0105] For example, in a city red-themed square scene, the initial design components consist of a sculpture, granite material, and a solemn design. First, the sculpture component parameters are extracted. Referring to the design specifications for red-themed commemorative sculptures and combining the attribute data of "sculpture" in the knowledge graph, its spatial dimensions are determined to be 8 meters high, with a base area of ​​4 meters x 4 meters, and a 0.5-meter-wide foundation space required for installation. Its light response characteristics are adapted to light intensity of 2500-8000 lux and color temperature of 4000-5500K, ensuring that the surface texture and details of the sculpture are clearly presented under different lighting conditions. This component has no sound interaction function, so the sound interaction parameter is empty. Next, the parameters for the supporting materials and processes are extracted. The parameters for the granite material need to match the sculpture installation requirements, determining that its single slab size is 1.2 meters x 1.2 meters, with a thickness of 0.15 meters, and its weather resistance parameters meet the standard for outdoor use for more than 15 years.

[0106] S1033. Verify the coordination between the component parameters and the environmental information, identify and eliminate mismatches between the component and the environment, and adaptively adjust the visual performance parameters and acoustic output parameters of the component.

[0107] In this embodiment of the application, firstly, a parameter matching degree calculation model is used to verify the coordination. This model is trained by collecting a large amount of environmental parameter and component parameter matching case data of public art design, with successful matching cases as positive samples and unmatched cases as negative samples, to obtain a classification model that can quickly identify unmatched situations.

[0108] Secondly, the matching degree between component parameters and environmental information is calculated using a formula:

[0109] ;

[0110] in, The range of matching degree values ​​is: , Indicates the number of parameter dimensions. Indicates the first The standardized values ​​of the component parameters, Indicates the first The standardized values ​​of each environmental information parameter are determined. Next, a matching threshold is set. Taking the scene adaptation requirements of the red-themed square as the standard, the matching degree threshold is set to 0.85. M≥0.85 is judged as a basic match, and M<0.85 is judged as a mismatch.

[0111] Subsequently, core parameters such as spatial dimensions, light intensity, color temperature, and sound output volume were selected. The standardized values ​​of the initial design components' parameters and the standardized values ​​of the collected environmental information were substituted into the formula for calculation, and mismatches were identified one by one. Finally, based on the type of mismatch, the visual performance parameters of the components, such as light response range and color adaptation standards, and the acoustic output parameters, such as volume and frequency, were adaptively adjusted to eliminate the mismatch.

[0112] Taking a red-themed urban plaza scene as an example, we first select two parameters: spatial size, light intensity, and color temperature. The standardized values ​​for the component parameters are 0.8, 0.7, and 0.6, respectively, and the standardized values ​​for the environmental information parameters are 0.9, 0.5, and 0.4, respectively. Then, we calculate the matching degree of the sculpture components. The calculation process is as follows: [Molecular part] denominator Matching degree .

[0113] Next, the matching threshold was set to 0.85, and the matching degree of the sculpture component was... The result was determined to be a basic match, requiring only minor adjustments. Finally, parameters were adaptively adjusted based on the mismatches. For the lighting mismatch of the sculpture, its lighting response characteristics were adjusted to an adaptation range of 2500-9000 lux to ensure that the details of the sculpture remain clear under strong midday light and that the solemn visual effect of the red theme is not affected. At the same time, the UV resistance parameters of the red theme coloring process were optimized to prevent fading caused by strong light.

[0114] S1034. Integrate the adjusted components to generate a target design component set that is adapted to the environment.

[0115] In one specific implementation, all adjusted components undergo secondary verification to confirm that the parameters of each component meet the environmental requirements and that there are no conflicts. After successful verification, all adjusted components are integrated and summarized to form the target design component set. The secondary verification adopts a cross-validation method, that is, by simulating environmental parameters to input the adjusted component parameters, it is verified whether the component output effect meets expectations, and at the same time, it checks whether there are parameter conflicts between different components, ensuring the environmental adaptability and stability of the target design component set.

[0116] Continuing with the example above, we first simulated the environmental parameters of the plaza at different times of day, verifying the adaptability of each adjusted component. For the sculpture, we input a midday light parameter of 8500 lux to verify whether its adjusted light response range of 2500-9000 lux could be properly adapted. Simultaneously, we checked whether the weather resistance parameters of the granite material matched the outdoor environment of the plaza, ensuring that the sculpture would not crack or weather over long-term use. Next, we checked for parameter conflicts between components: we tested the distance between the sculpture's installation area and the relief display wall, simulating the acoustic effects of both operating simultaneously. We confirmed that the total volume of the relief's audio narration and the ambient sound effects around the sculpture did not exceed 65 decibels, thus not disrupting the solemn atmosphere of the plaza. We also verified the installation space dimensions, ensuring that the 4m x 4m footprint of the sculpture perfectly matched the reserved base, and that the 15m x 3m size of the relief matched the space on the south wall, with no spatial conflicts. After this second verification, all component parameters met the environmental compatibility requirements and there were no mutual conflicts. The final integration and summarization form a set of target design components, which can directly support the implementation of public art in urban red-themed squares.

[0117] This application solves the problem of traditional design where components are disconnected from the physical environment and artistic effects cannot be properly presented due to environmental influences by accurately acquiring spatial environmental information and quantitatively matching and adjusting component parameters. Compared with the traditional adaptation method that relies on experience judgment, it improves the accuracy and stability of the adaptation between design components and the environment, ensuring that public art can play its best role in actual space.

[0118] S104. Organize and sort the components in the target design component set according to the preset plot development logic to form a narrative structure sequence;

[0119] In this step, the plot development logic refers to the content progression of the public art experience process, including stages such as introduction, development, climax, and conclusion, used to guide the audience to perceive the spatial cultural connotation. The narrative structure sequence is the sequence of artistic expression execution built upon the plot development logic, specifying the temporal and state transition relationships of different artistic expressions in the experience process.

[0120] Optionally, such as Figure 2 As shown, step S104 may specifically include the following steps:

[0121] S1041. Set the plot development logic, define the corresponding theme and emotional tone for each stage of the plot development logic, and classify the components in the target design component set according to their artistic expression.

[0122] In one specific implementation, the plot development logic is first set by combining the cultural and emotional positioning of the target public space, and divided into three stages: cultural origin, emotional resonance and value sublimation. For example, the themes are defined as historical imprint, urban warmth and future outlook, and the corresponding emotional tones are profound, warm and uplifting.

[0123] Subsequently, the components in the target design component set are categorized according to their artistic expression: sculpture installations, interactive lighting, and digital imaging. For example, in the target design component set for the city's red-themed plaza, traditional relief sculptures are classified as sculpture installations, gradient light and shadow lights as interactive lighting, and urban development projections as digital imaging.

[0124] S1042. Based on the theme and emotional tone, select matching components from the classified components, and determine the display order and duration within the corresponding plot stage according to the attributes of each component, so as to generate the component sequence corresponding to each stage.

[0125] Specifically, matching components are first selected based on the theme and emotional tone of each stage. For example, traditional relief sculptures and historical scene projections are selected for the cultural origin stage, interactive lighting and citizen story videos are selected for the emotional resonance stage, and future-themed sculptures and gradient light and shadow lights are selected for the value sublimation stage.

[0126] Next, the display order and duration are determined based on the components' visual impact, information delivery efficiency, and other attributes. Specifically, the priority of component display is determined using a weighted scoring method, with visual impact weighted at 0.4, information delivery efficiency weighted at 0.3, and audience interaction threshold weighted at 0.3. The order is determined by calculating the overall score of the components, with components scoring higher being displayed first. The duration is set based on the component's content capacity and the audience's attention span, with static display components lasting no less than 8 minutes and dynamic interactive components lasting no less than 12 minutes, ensuring that the audience can fully receive the cultural information.

[0127] For example, sculptures and reliefs are selected for the historical commemoration stage to match the solemn and dignified emotional tone; interactive commemorative installations and red story projections are selected for the spiritual inheritance stage to match the respectful and remembrance emotional tone; and themed slogans and light signs and dynamic light and shadow corridors are selected for the era's progress stage to match the uplifting emotional tone.

[0128] The components at each stage were weighted and scored. Taking the historical commemoration stage as an example, the sculpture scored 9 points for visual impact, 8 points for information delivery efficiency, and 7 points for audience interaction threshold, for a total score of 9×0.4+8×0.3+7×0.3=8.1; the relief scored 7 points for visual impact, 9 points for information delivery efficiency, and 8 points for audience interaction threshold, for a total score of 7×0.4+9×0.3+8×0.3=7.9. ​​Because the sculpture scored higher overall, it was decided to display the sculpture first, followed by the relief, in this stage.

[0129] The sculptures are static display components, with a duration of 10 minutes based on their rich detail; the reliefs, accompanied by audio narration, have a larger content capacity and a duration of 12 minutes. Similarly, the interactive commemorative installations for the spiritual inheritance phase are dynamic interactive components, with a duration of 15 minutes, and the red story projection is set for a duration of 12 minutes; the theme slogan light boards for the era of striving forward are set for a duration of 8 minutes, and the dynamic light and shadow corridor is set for a duration of 15 minutes.

[0130] The final product consists of a sequence of components for each stage: the historical commemoration stage includes a 10-minute sculpture and a 12-minute relief; the spiritual inheritance stage includes a 15-minute interactive commemorative installation and a 12-minute projection of revolutionary stories; and the era-striving stage includes an 8-minute themed slogan light board and a 15-minute dynamic light and shadow corridor.

[0131] S1043. Connect the component sequences corresponding to each stage according to the chronological order of the plot development logic, and configure state transition rules at the junction of adjacent sequences to form a narrative structure sequence.

[0132] Specifically, step S1043 may include the following processes: extracting the start and end time markers of the component sequences corresponding to each stage; determining the order of each sequence according to the plot development logic; arranging them sequentially on the timeline; analyzing the attribute differences between the end and start components in adjacent sequences; calculating the intermediate attribute value sequence required for smooth transition based on the attribute differences; inserting state transition instructions at the connection points of adjacent sequences; and synthesizing the chronologically arranged sequences with the state transition instructions to generate a narrative structure sequence.

[0133] Among them, the state transition rule is a set of parameters used to standardize the connection method of adjacent component sequences, including intermediate attribute value sequence transition time length conversion instructions, etc., to achieve smooth switching of artistic expression forms; the state transition instruction includes the intermediate attribute value sequence and the corresponding transition time length.

[0134] In one specific implementation, firstly, the start and end time markers of each stage component sequence are extracted, the sequence order is determined according to the plot development logic, and then all sequences are arranged sequentially along the timeline to clarify the connection nodes between sequences.

[0135] Next, the core attribute differences between the end and beginning components of adjacent sequences are analyzed, and attribute dimensions that have a significant impact on visual or auditory performance are selected. An interpolation algorithm is used to calculate the intermediate attribute value sequence required for a smooth transition. By using uniform or variable-speed attribute changes, abrupt changes in performance during the transition are avoided.

[0136] Based on attribute differences, the sequence of intermediate attribute values ​​required for a smooth transition is calculated using a linear interpolation formula:

[0137] ;

[0138] in, Indicates the first step in the transition process The attribute value at time. Indicates the initial attribute value. Indicates the target attribute value. Indicates the total transition time. Indicates a transition point in time.

[0139] The core advantage of linear interpolation formulas is that they achieve a smooth and uniform change in attribute values, avoiding visual discontinuities caused by abrupt changes in attributes during the transition. If a more rhythmic transition effect is required, non-linear interpolation formulas, such as quadratic interpolation formulas, can be used instead, and the acceleration or deceleration of attribute value changes can be achieved by adjusting the interpolation curve.

[0140] Then, at the junctions of adjacent sequences, state transition instructions containing intermediate attribute value sequences and transition time lengths are inserted, along with an exception handling mechanism to ensure that a single component failure does not affect the overall process. Finally, the chronologically ordered sequence of components is integrated with the state transition instructions at the junctions to generate a complete and smoothly transitioning narrative structure sequence.

[0141] For example, in the scenario of a city's red-themed square, the three-stage component sequence and time markers of the public art display in the city's red-themed square are first clearly defined: the historical remembrance stage sequence, ending at minute 22; the spiritual inheritance stage sequence, starting at minute 22 and ending at minute 49; and the contemporary progress stage sequence, starting at minute 49. Based on the logic of plot development, the sequence order is determined as historical remembrance, spiritual inheritance, and contemporary progress, and arranged in this order on the timeline.

[0142] Next, brightness and color saturation, two core visual attributes, were selected to analyze the attribute differences between adjacent sequences. The brightness of the relief sculpture at the end of the historical commemoration phase was 100 cd / m², and the color saturation was 0.6. The brightness of the interactive commemorative device at the beginning of the spiritual inheritance phase was 200 cd / m², and the color saturation was 0.8. A total transition time was set. The intermediate value sequence for brightness is 125 cd / m² at 0.5 minutes, 150 cd / m² at 1 minute, 175 cd / m² at 1.5 minutes, and 200 cd / m² at 2 minutes. The intermediate value sequence for color saturation is 0.65 at 0.5 minutes, 0.7 at 1 minute, 0.75 at 1.5 minutes, and 0.8 at 2 minutes.

[0143] Then, a state transition instruction is inserted at the junction of adjacent sequences. This instruction includes the aforementioned intermediate attribute value sequence and a transition time of 2 minutes. Finally, the chronologically ordered sequence and the state transition instruction are combined to generate a complete narrative structure sequence. In another specific implementation, a nonlinear interpolation formula can be used to calculate the intermediate attribute value sequence to achieve a more artistic transition effect; this application does not limit this approach.

[0144] State transition instructions must be synchronously written into the component's control parameters. For example, a brightness increase instruction should be triggered at the end of the historical scene projection, and a color saturation adjustment instruction should be synchronously executed at the beginning of the citizen story video to ensure a seamless transition between the two components. At the same time, the instructions must include an exception interruption handling mechanism to prevent a single component failure from affecting the overall narrative flow.

[0145] This application solves the problem of disordered and fragmented experience flow in the display of components in traditional public art design by constructing a logical plot development and configuring smooth state transition rules. Compared with the traditional unordered display method, it realizes the continuity and hierarchy of the art experience and enhances the audience's cultural perception and emotional engagement.

[0146] S105. Based on the artistic expressions in the narrative structure sequence, virtual scene data associated with real spatial locations is generated through deconstruction and parametric transformation, and augmented reality fusion rendering is performed on the virtual scene data to generate an art experience scheme.

[0147] Parametric transformation is the process of converting the visual motion and other characteristics of artistic expressions into quantifiable and adjustable parametric models.

[0148] Virtual scene data is a collection of digital data containing information such as the spatial coordinates and time stamps of virtual elements, used to present artistic content in augmented reality environments.

[0149] Augmented reality blending rendering is a technical process that adjusts the presentation of virtual scene data to make it blend naturally with the real environment.

[0150] Optionally, step S105 may specifically include the following steps:

[0151] S1051. Decompose the artistic expression in the narrative structure sequence into virtual elements, the virtual elements including graphics, color values, motion trajectories and time markers, and convert the virtual elements into parametric models;

[0152] Among them, the parametric model is a digital model that uses specific parameters to describe the attributes of virtual elements. The parameters cover graphic size, color value, motion speed, time nodes, etc., which facilitates precise adjustment and spatial mapping.

[0153] In one specific implementation, each artistic expression in the narrative structure sequence is first traversed and broken down into four basic virtual elements: graphics, color values, motion trajectories, and time markers. Static artistic expressions do not have motion trajectories, while dynamic artistic expressions require explicit details of their motion trajectories.

[0154] Then, for each type of virtual element after decomposition, attribute quantification is performed: the graphic shape is converted into quantifiable size values, such as the length and width of a rectangle, the radius of a circle, and the coordinates of key vertices of irregular shapes; standardized color modes such as RGB and HSV are used to convert colors into specific values ​​to ensure accurate color reproduction; dynamic elements with motion trajectories need to quantify values ​​such as starting point coordinates, ending point coordinates, motion speed, and motion cycle; the time stamp parameters need to clearly define the start and end times of the narrative timeline corresponding to each virtual element to ensure matching with the overall narrative rhythm.

[0155] Finally, the quantified parameters are integrated and linked hierarchically according to artistic expression, virtual elements, and quantified parameters to form a parametric model that can be directly used for digital production.

[0156] For example, in a city red-themed square scene, two core artistic expressions are selected for deconstruction and parametric modeling.

[0157] Sculpture is a static art form, corresponding to a narrative sequence of 0 to 10 minutes. It is first broken down into four categories of virtual elements: graphics, color values, motion trajectories, and time markers. The graphics are irregular human figure outlines, which can be simplified to a core rectangular body with a triangular top decoration. The color value is the natural color of granite, and there is no motion trajectory. The time markers are 0 to 10 minutes. Next, the attributes of the virtual elements are quantified: the rectangular body is 4 meters long, 4 meters wide, and 8 meters high; the triangular top decoration has a base length of 4 meters and a height of 1.2 meters; the color values ​​are quantized using the RGB color model as 160, 160, 160; the motion trajectory parameter is left blank; and the time marker parameters clearly specify a start time of 0 minutes and an end time of 10 minutes. Then, these quantified parameters are integrated according to the logical relationships between artistic expression, virtual elements, and parameter values ​​to form a parametric model of the sculpture. This model is presented as a digital document that can be directly imported into 3D modeling software. After inputting the parameters, the basic three-dimensional outline of the sculpture can be generated, and the length, width, and height values ​​can be adjusted to adapt to different installation spaces.

[0158] The dynamic light and shadow corridor is a dynamic art form corresponding to a narrative sequence of 61 to 76 minutes. It is broken down into four virtual elements: graphics, color values, movement trajectories, and time markers. The graphics are continuously flowing strips of light and shadow; the color values ​​are a red-gold gradient; the movement trajectory is the strips of light and shadow flowing uniformly from the north side of the square to the south side; and the time markers are for minutes 61 to 76. Next, attribute quantification is performed: each strip of light and shadow is 5 meters long and 0.3 meters wide, with a 1-meter spacing between adjacent strips; the color values ​​use the RGB color model, with the initial red being 255,0,0 and the transitioning gold being 255,215,0; the starting point coordinates are 0 meters, 30 meters from the north edge of the square, and the ending point coordinates are 0 meters, 0 meters from the south edge of the square; the movement speed is 2 meters per minute; and the time marker parameters clearly specify the start time as 61 minutes and the end time as 76 minutes. These parameters are then integrated to form a parametric model of the dynamic light and shadow corridor. This model is a set of light and shadow control parameters that can be run directly. After being imported into the lighting control equipment, the equipment can control the flow direction and speed of light and shadow according to the parameters. Adjusting the color value parameters can switch the gradient effect, and adjusting the time parameters can match the overall narrative rhythm.

[0159] S1052. Associate the parameterized model with the real spatial location, and use spatial anchoring technology to map the position of the virtual element to a specific point in the corresponding physical space to generate virtual scene data with spatial association.

[0160] Specifically, step S1052 may include the following processes: Based on the position parameters in the parameterized model and the coordinate data of the real space, select multiple physical points in the real space as spatial anchor points, and mark the physical coordinates and corresponding virtual coordinates of the spatial anchor points; based on the physical coordinates and virtual coordinates, calculate coordinate transformation parameters, establish a transformation relationship from the virtual coordinate system to the real coordinate system, the transformation relationship including a rotation matrix and a translation vector; apply the transformation relationship to transform the coordinates of the virtual element in the parameterized model, obtain the corresponding coordinates of the virtual element in the real space, and adjust the position of the virtual element to match a specific point in the physical space; group all the virtual elements with adjusted positions according to spatial proximity to generate a group of virtual elements with spatial association.

[0161] In the above steps, spatial anchoring technology establishes a correspondence between virtual coordinates and real coordinates by setting fixed physical points in real space as anchors, thereby enabling the spatial positioning of virtual elements. Spatial anchors are reference points with definite physical coordinates in real space, used to precisely anchor the position of virtual elements.

[0162] In this embodiment of the application, firstly, based on the position parameters in the parameterized model and the coordinate data of the real space, multiple points with clear physical identifiers are selected in the real space as spatial anchor points. Each anchor point is labeled with unique physical coordinates and corresponding virtual coordinates, and a virtual-real correspondence relationship is established at the anchor point level.

[0163] Next, coordinate transformation parameters are calculated based on physical and virtual coordinates to establish a transformation relationship from the virtual coordinate system to the real coordinate system. This transformation relationship includes rotation matrices and translation vectors. The least squares method is used to calculate the transformation parameters, assuming the virtual coordinates are... The actual coordinates are The conversion formula is:

[0164] ;

[0165] in, for Rotation matrix, Translation vector .

[0166] Then, the virtual coordinates of all virtual elements in the parameterized model are substituted into the established transformation relationship to calculate the corresponding coordinates of each virtual element in the real space; the position of the virtual element is adjusted according to the calculation results to make it accurately match the specific installation or display point in the real physical space.

[0167] Finally, the K-means clustering algorithm can be used, with the real spatial coordinates of virtual elements as feature values. The number of clusters is set according to the number of functional partitions in the real space, and virtual elements that are close in location are grouped together. All grouped virtual elements and their real-time spatial coordinates are integrated to generate virtual scene data with clear spatial relationships, which facilitates subsequent rendering control by region.

[0168] For example, in the scenario of a red-themed urban plaza, the actual spatial coordinates of the plaza are first measured. Combined with the virtual coordinate range in the parametric model, fixed landscape stones at the four corners of the plaza are selected as spatial anchor points, and their physical coordinates and corresponding virtual coordinates are marked. Specifically: Anchor point 1: physical coordinates (0m, 0m, 0m), virtual coordinates (0,0,0); Anchor point 2: physical coordinates (90m, 0m, 0m), virtual coordinates (90,0,0); Anchor point 3: physical coordinates (90m, 60m, 0m), virtual coordinates (90,60,0); Anchor point 4: physical coordinates (0m, 60m, 0m), virtual coordinates (0,60,0).

[0169] Next, using the virtual and real coordinates of the four anchor points as the basic data, a system of equations is constructed using the least squares method to solve for the transformation parameters. Substituting the coordinate data of the four anchor points into the system of equations, the rotation matrix is ​​obtained. Translation vector This means that the virtual coordinates correspond to the real coordinates in a 1:1 ratio, without the need for additional angle correction or position offset.

[0170] Then, the virtual coordinates of each parametric model are extracted and substituted into the transformation formula to calculate the real coordinates. For example, the virtual coordinates of the sculpture (20,30,0) are transformed into real coordinates of (20m,30m,0m), corresponding to the reserved base position in the central memorial area of ​​the square; the virtual coordinates of the relief (70,30,0) are transformed into real coordinates of (70m,30m,0m), corresponding to the relief display wall position on the south side of the square; the virtual coordinates of the dynamic light and shadow corridor (45,10,0) to (45,50,0) are transformed into real coordinates of (45m,10m,0m) to (45m,50m,0m), corresponding to the central passage area in the north-south direction of the square. The positions of each virtual element are adjusted to perfectly match these physical points.

[0171] Finally, scene data was generated by grouping according to spatial proximity: the real space of the square is divided into four functional areas: the central memorial area, the south relief area, the north interactive area, and the green areas on the east and west sides. Therefore, the number of K-means clusters was set to 4. Clustering was performed using the real coordinates of each virtual element as feature values. The sculpture was finally grouped into the central memorial area group, the relief into the south relief area group, the interactive memorial device into the north interactive area group, and the associated virtual elements of the dynamic light and shadow corridor into the central memorial area extension group. All grouped virtual elements and their real-time spatial coordinates were integrated to generate virtual scene data with spatial associations. Subsequently, the AR rendering timing and effects of each virtual element can be precisely controlled by region.

[0172] S1053. According to augmented reality rendering rules, adjust the transparency, shadows, and lighting effects of the virtual scene data to blend with the real environment;

[0173] In one specific implementation, augmented reality rendering rules are set with reference to ambient lighting and object reflection in real space, and parameters are adjusted for the characteristics of different virtual elements.

[0174] For example, the virtual elements of traditional relief sculptures are set with a transparency of 0.9. To make them appear more realistic, the shadow angle is set to 45 degrees based on the direction of the midday sun in the square, and the shadow length is 0.5 times the height of the sculpture. The intensity of the light reflection is also adjusted to match the reflection effect of the surrounding objects. For interactive lighting virtual elements, the transparency is set to 0.7 to avoid them being too abrupt. The brightness of the self-illumination is adjusted according to the ambient light intensity so that they can be clearly displayed without being dazzling during the day and at night.

[0175] Augmented reality rendering can use game engines such as Unity and Unreal Engine. By writing shaders, the lighting and shadow effects of virtual elements can be adjusted. The shader program can read the lighting sensor data of the real environment in real time and dynamically adjust the brightness and shadow parameters of the virtual elements to achieve lighting consistency between the virtual and real environments.

[0176] S1054. Based on the timeline of the narrative structure sequence, the rendering output of the virtual scene data is triggered sequentially to generate an art experience scheme.

[0177] The art experience solution includes time-stamped virtual content playback instructions.

[0178] In one specific implementation, the timeline information in the narrative structure sequence is first extracted, and the rendering output logic of the virtual scene data is arranged in chronological order. For example, at minute 0, the rendering output of the virtual elements of the traditional relief sculpture is triggered; at minute 10, the rendering output of the virtual elements of the historical scene projection is triggered; at minute 18, the rendering output of the virtual elements of the citizen story video is triggered through a state transition command, and so on.

[0179] Then, these rendering output logics are organized into playback instructions containing time stamps and corresponding virtual content, integrating them into a complete art experience solution. Users can then see the merged art content in the corresponding locations in the real space according to the time sequence outlined in the solution using augmented reality devices. The art experience solution must include device compatibility specifications, such as supporting different terminal devices like mobile AR and AR glasses, adjusting the rendering precision of virtual elements for different device resolutions and field of view. The solution must also include a debugging mode, allowing on-site personnel to adjust the position and parameters of virtual elements in real time to cope with sudden environmental changes in the real space.

[0180] This application achieves a precise connection between virtual and real spaces through parametric transformation and spatial anchoring technology. Combined with augmented reality fusion rendering, it solves the problems of monotonous forms and insufficient audience interaction in traditional public art. Compared with the traditional static display mode, it enriches the artistic expression and enhances the audience's immersive experience and sense of participation.

[0181] Figure 3 This is a schematic diagram illustrating a specific implementation of a knowledge graph-based public art design scheme generation system provided in this application. (Refer to...) Figure 3 The system may include:

[0182] Module 31 is used to construct a knowledge graph based on design element information in the field of public art. The knowledge graph includes entities that express design intentions and semantic relationships that connect the entities.

[0183] Matching module 32 is used to acquire text content reflecting the cultural sentiment of the target public space, identify sentiment information from the text content, and associate and match the sentiment information with entities expressing design intent in the knowledge graph, and determine the initial design component set from the knowledge graph based on the matching results;

[0184] Processing module 33 is used to perform environmental adaptability processing on the elements in the initial design component set based on the physical layout, ambient lighting and acoustic characteristics of the target public space, and to determine the target design component set.

[0185] The sorting module 34 is used to organize and sort the components in the target design component set according to the preset plot development logic to form a narrative structure sequence, wherein the narrative structure sequence specifies the temporal and state transition relationships of different artistic expressions in the experience process;

[0186] The generation module 35 is used to generate virtual scene data associated with real spatial locations based on the artistic expressions in the narrative structure sequence through deconstruction and parametric transformation, and to perform augmented reality fusion rendering processing on the virtual scene data to generate an art experience scheme.

[0187] The knowledge graph-based public art design scheme generation system of this application is used to implement the aforementioned knowledge graph-based public art design scheme generation method. Therefore, the specific implementation of the knowledge graph-based public art design scheme generation system can be found in the embodiment section of the knowledge graph-based public art design scheme generation method above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0188] This application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of any of the above-described knowledge graph-based public art design scheme generation methods.

[0189] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the above-described knowledge graph-based public art design scheme generation methods.

[0190] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0191] Embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above embodiments of the knowledge graph-based public art design scheme generation method.

[0192] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0193] The foregoing has provided a detailed description of a knowledge graph-based public art design scheme generation method and system provided in this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are merely for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from its principles, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A method for generating public art design schemes based on knowledge graphs, characterized in that, include: A knowledge graph is constructed based on design element information in the field of public art. The knowledge graph includes entities that express design intentions and semantic relationships that connect the entities. The process involves acquiring textual content reflecting the cultural sentiment of the target public space, identifying sentiment information from the textual content, and matching this sentiment information with entities expressing design intent in the knowledge graph. Based on the matching results, an initial set of design components is determined from the knowledge graph, including: The process involves: acquiring text content from a target public space; extracting words with cultural characteristics and emotional states from the text content to form vocabulary groups; assigning emotional values ​​to the words in the vocabulary groups; scoring the words in the vocabulary groups across multiple dimensions; generating sentiment tendency information by combining all scoring results, which includes sentiment vectors; labeling entities expressing design intent with sentiment attributes in the knowledge graph; calculating the matching degree between the sentiment vectors and the sentiment attributes; selecting matching entities based on the matching degree; using the matching entities as candidate components; and adding closely related adjacent entities to the candidate components based on the relationships between entities in the knowledge graph to form an initial set of design components. Based on the physical layout, ambient lighting, and acoustic characteristics of the target public space, environmental adaptability processing is performed on the elements in the initial design component set to determine the target design component set. According to the preset plot development logic, the components in the target design component set are organized and sorted to form a narrative structure sequence, which specifies the temporal and state transition relationships of different artistic expressions in the experience process; Based on the artistic expressions in the narrative structure sequence, virtual scene data associated with real spatial locations is generated through deconstruction and parametric transformation. Augmented reality fusion rendering is then performed on the virtual scene data to generate an artistic experience scheme.

2. The method according to claim 1, characterized in that, According to the preset plot development logic, the elements in the target design component set are organized and sorted to form a narrative structure sequence, including: Define the plot development logic, define the corresponding theme and emotional tone for each stage of the plot development logic, and classify the components in the target design component set according to their artistic expression. Based on the theme and emotional tone, matching components are selected from the categorized components, and the display order and duration within the corresponding plot stage are determined according to the attributes of each component, so as to generate the component sequence corresponding to each stage. The component sequences corresponding to each stage are connected according to the chronological order of the plot development logic, and state transition rules are configured at the junctions of adjacent sequences to form a narrative structure sequence.

3. The method according to claim 2, characterized in that, The component sequences corresponding to each stage are connected according to the chronological order of the plot development logic, and state transition rules are configured at the junctions of adjacent sequences to form a narrative structure sequence, including: Extract the start and end time markers of the component sequences corresponding to each stage, determine the order of each sequence according to the plot development logic, and arrange them sequentially on the timeline; Analyze the attribute differences between the last component and the first component in adjacent sequences, and calculate the intermediate attribute value sequence required for a smooth transition based on the attribute differences; Insert a state transition instruction at the junction of adjacent sequences, wherein the state transition instruction includes the intermediate attribute value sequence and the corresponding transition time length; The chronologically ordered sequence is combined with the state transition instructions to generate a narrative structure sequence.

4. The method according to claim 1, characterized in that, Based on the artistic expressions within the narrative structure sequence, virtual scene data associated with real-world spatial locations is generated through deconstruction and parametric transformation. Augmented reality fusion rendering is then performed on this virtual scene data to generate an artistic experience scheme, including: The artistic expressions in the narrative structure sequence are decomposed into virtual elements, which include graphics, color values, motion trajectories and time markers, and the virtual elements are converted into parametric models. The parameterized model is associated with the real spatial location, and the position of the virtual element is mapped to a specific point in the corresponding physical space through spatial anchoring technology, thereby generating virtual scene data with spatial association. Based on augmented reality rendering rules, the transparency, shadows, and lighting effects of the virtual scene data are adjusted to blend with the real environment; Based on the timeline of the narrative structure sequence, the rendering output of the virtual scene data is triggered sequentially to generate an art experience scheme, which includes virtual content playback instructions with time stamps.

5. The method according to claim 4, characterized in that, The parameterized model is associated with real spatial locations. Using spatial anchoring technology, the positions of the virtual elements are mapped to specific points in the corresponding physical space, generating a group of virtual elements with spatial associations, including: Based on the position parameters in the parameterized model and the coordinate data of the real space, multiple physical points are selected in the real space as spatial anchor points, and the physical coordinates and corresponding virtual coordinates are marked for the spatial anchor points. Based on the physical coordinates and virtual coordinates, coordinate transformation parameters are calculated, and a transformation relationship from the virtual coordinate system to the real coordinate system is established. The transformation relationship includes a rotation matrix and a translation vector. By applying the transformation relationship, the coordinates of the virtual element in the parameterized model are transformed to obtain the corresponding coordinates of the virtual element in the real space, and the position of the virtual element is adjusted to match a specific point in the physical space. All the virtual elements with adjusted positions are grouped according to spatial proximity to generate virtual element groups with spatial relationships.

6. The method according to claim 1, characterized in that, Based on the physical layout, ambient lighting, and acoustic characteristics of the target public space, environmental adaptability processing is performed on the elements in the initial design component set to determine the target design component set, including: Acquire environmental information of the target public space, including physical layout, ambient lighting and acoustic characteristics; Component parameters are extracted from the initial design component set, including spatial size requirements, lighting response characteristics, and sound interaction parameters; The component parameters are verified to be compatible with the environmental information, mismatches between the components and the environment are identified and eliminated, and the visual performance parameters and acoustic output parameters of the components are adaptively adjusted. The adjusted components are integrated to generate a set of target design components that are adapted to the environment.

7. A public art design scheme generation system based on knowledge graphs, used to execute the public art design scheme generation method based on knowledge graphs as described in any one of claims 1 to 6, characterized in that, include: A construction module is used to build a knowledge graph based on design element information in the field of public art. The knowledge graph contains entities that express design intentions and semantic relationships that connect the entities. The matching module is used to acquire text content that reflects the cultural sentiment of the target public space, identify sentiment information from the text content, and associate and match the sentiment information with entities in the knowledge graph that express design intentions. Based on the matching results, the initial design component set is determined from the knowledge graph. The processing module is used to perform environmental adaptability processing on the elements in the initial design component set based on the physical layout, ambient lighting and acoustic characteristics of the target public space, and to determine the target design component set. The sorting module is used to organize and sort the components in the target design component set according to the preset plot development logic to form a narrative structure sequence, wherein the narrative structure sequence specifies the temporal and state transition relationships of different artistic expressions in the experience process; The generation module is used to generate virtual scene data associated with real spatial locations based on the artistic expressions in the narrative structure sequence through deconstruction and parametric transformation, and to perform augmented reality fusion rendering processing on the virtual scene data to generate an art experience scheme.

8. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the steps of the knowledge graph-based public art design scheme generation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables the generation method of public art design schemes based on knowledge graphs as described in any one of claims 1 to 6.

Images