An automated method and system for architectural design scenarios

Abstract

The present application relates to the field of three-dimensional scene generation, and more particularly to a kind of automatic generation method and system for architectural design scene, the present application sets up label set for model body, for the model body in model library is taken sight switching verification, and real-time record several rendering attribute parameters in the process of sight switching verification, to determine the correlation characteristic value between label set based on the change characteristics of rendering attribute parameters in time domain dimension, to determine the correlation, subsequent in user end complete model body deployment, capture the current view range and moving direction of viewfinder, based on the adjacent view range predicted based on moving direction, based on the correlation between model body in current view range relative adjacent view range model body is rendered.The present application is adapted to render for three-dimensional model in different scene space in advance through sight switching verification, reduce the frame freezing or rendering exception of rendering picture when viewfinder moves in three-dimensional space, improve picture smoothness.

Description

Technical Field

[0001] This invention relates to the field of 3D scene generation, and more particularly to an automated generation method and system for architectural design scenes. Background Technology

[0002] With the rapid development of architectural visualization technology, scene construction and rendering have become core aspects of architectural design expression. For example, manually building scenes using 3D modeling software provides a new path for constructing large-scale and complex architectural scenes. Based on this, diverse scene layouts such as urban blocks and indoor spaces can be quickly generated for users to browse scenes from multiple perspectives.

[0003] For example, Chinese Patent Publication No. CN117333349A discloses an image rendering method, system, device, and storage medium in a design scene. The method includes the following steps: generating a model scene diagram; receiving at least one camera setting parameter input by a user; receiving a screen generation instruction and generating a corresponding camera display screen based on the camera selected by the user and the at least one camera setting parameter input by the user; receiving a screen rendering instruction and generating a rendered image based on the selected camera display screen. This automatic rendering software, applied to industrial design, achieves automatic image rendering effects in design scenes, improving design efficiency.

[0004] However, the following problems still exist in the existing technology. Due to the high randomness and complexity of the design scenarios, and the frequent switching of user perspectives during the application process, in some cases, the difference in scene can easily lead to problems such as screen stuttering and rendering abnormalities when switching viewpoints, which affects the display effect and reduces the user experience. Summary of the Invention

[0005] To address this, the present invention provides an automated generation method and system for architectural design scenarios, which overcomes the problems in the prior art where the design scenarios are highly random and complex, and users frequently switch perspectives to observe the design scenarios during application. In some cases, when switching viewpoints, the differences in scenarios can easily lead to problems such as screen stuttering and rendering abnormalities, affecting the display effect and reducing the user experience.

[0006] To achieve the above objectives, in one aspect, the present invention provides an automated generation method for architectural design scenarios, comprising, Set a tag set for several model objects in the model library. The tag set includes a main tag to identify the model object and several sub-tags that reflect the range of the three-dimensional attributes of the model object. The framing switching verification for the model objects in the model library includes grouping the model objects corresponding to different tag sets into a predetermined three-dimensional space, and controlling the viewfinder to switch the framing range between the model objects. Several rendering attribute parameters are recorded in real time during the framing switching verification process. Based on the change characteristics of the rendering attribute parameters in the temporal dimension, the correlation feature values ​​between the tag sets are determined to determine the correlation relationship. In response to the completion of model deployment on the user side, the current framing range and movement direction of the viewfinder are captured, and the adjacent framing range is predicted based on the movement direction. Determine the tag sets corresponding to the model objects within the current framing range relative to adjacent framing ranges to establish relationships. Then, render the model objects based on these relationships, including... The location distribution type is determined based on the spatial coordinates of the model objects within the adjacent framing range, and the pre-rendering strategy is selected based on the location distribution relationship; Real-time rendering of the model within the adjacent framing range; The pre-rendering strategy includes global pre-rendering or determining local clustered model bodies, and performing local pre-rendering on all model bodies in the region where the local clustered model body is located.

[0007] Furthermore, the process of setting tag sets for several model bodies within the model library includes, Determine the type number corresponding to the type of the model body as the main label; Determine several three-dimensional attributes of the model body, and determine the range of the three-dimensional attributes of the model body corresponding to each of the three-dimensional attributes of the model body; The three-dimensional attribute range of each model body is used as a sub-label of the model body; Each tag set corresponds one-to-one with a model body, and the three-dimensional attributes of the model body are preset.

[0008] Furthermore, during the process of grouping the model bodies into a predetermined three-dimensional space, Two model objects are placed at the same time, and the distance between the model objects must be greater than half of the viewfinder's field of view, and the label sets of the model objects placed at the same time must be different.

[0009] Furthermore, the process of switching the framing range between the various model bodies by controlling the viewfinder includes, Control the viewfinder to correspond to any model body, so that the initial framing range contains only a single model body; Move the viewfinder until the viewfinder contains only the complete other model body described above; Move the viewfinder again until it corresponds to the initial framing range.

[0010] Furthermore, the process of determining the association features between tag sets based on the variation characteristics of rendering attribute parameters in the temporal dimension, in order to determine the association relationship, includes: Record rendering attribute parameters, construct temporal variation curves of rendering attribute parameters, determine the amplitude difference ratio between curve segments corresponding to when the view range includes the model body, and determine the amplitude difference ratio as the associated feature value; If the associated feature value is greater than a predetermined associated threshold, it is determined that there is an association between the tag sets. The rendering attribute parameter is either GPU utilization or single-frame rendering time.

[0011] Furthermore, rendering the model body based on the association between the corresponding tag sets of the model body within the current framing range and the adjacent framing range includes... If there is a correlation between the current field of view and the corresponding tag set of the model body in the adjacent field of view, the position distribution type is determined based on the spatial coordinates of the model body in the adjacent field of view, and the pre-rendering strategy is selected based on the position distribution relationship; If there is no correlation between the label sets of the model objects in the current field of view and those in the adjacent field of view, then the model objects in the adjacent field of view will be rendered in real time.

[0012] Furthermore, the process of determining the location distribution type based on the spatial coordinates of the model volume within the adjacent framing range includes, Determine the spatial coordinates of model bodies within the adjacent framing range that have related label sets; Determine the distances between the spatial coordinates and their nearest neighbor spatial coordinates to solve for the mean distance; If the mean distance is greater than or equal to the distance concentration threshold, it is determined to be a dispersed distribution type; If the mean distance is less than the distance concentration threshold, it is determined to be a concentrated distribution type.

[0013] Furthermore, selecting a pre-rendering strategy based on location distribution relationships includes, For distributed and scattered data types, global pre-rendering is used; For the concentrated and dispersed types, a local cluster model body is determined, and local pre-rendering is performed on all model bodies in the region where the local cluster model body is located.

[0014] Furthermore, the process of determining the local cluster model volume and performing local pre-rendering on all model volumes within the region where the local cluster model volume is located includes, Based on spatial coordinates, cluster the model bodies that have a relationship with the label set. If the clustering conditions are met, determine the cluster and the model body corresponding to the cluster is determined as the local cluster model body. The field of view is divided into several regions, and all models within the region where the local clustered model is located are pre-rendered locally. The clustering conditions are that the distance between the spatial coordinates of any model body within a cluster is less than a predetermined distance threshold, and the number of model bodies in a cluster must be greater than a predetermined number threshold.

[0015] On the other hand, a system that provides automated generation methods for architectural design scenarios includes: The tag segmentation module is used to set tag sets for several model bodies in the model library. The tag set includes a main tag to identify the model body and several sub-tags that reflect the range of the three-dimensional attributes of the model body. The verification module verifies the viewfinder switching of the model objects in the model library, including grouping the model objects corresponding to different tag sets into a predetermined three-dimensional space, and controlling the viewfinder to switch the viewfinder between the model objects. The association determination module is used to record several rendering attribute parameters in real time during the framing switching verification process, and determine the association feature values ​​between the tag sets based on the change characteristics of the rendering attribute parameters in the temporal dimension, so as to determine the association relationship. The range determination module responds to the user terminal to complete the deployment of the model body, captures the current framing range and movement direction of the viewfinder, and predicts the adjacent framing range based on the movement direction. The rendering control module is used to determine the corresponding tag sets of model objects within the current framing range relative to adjacent framing ranges, to determine the association relationship, and to render the model objects based on the association relationship between the tag sets, including... The location distribution type is determined based on the spatial coordinates of the model objects within the adjacent framing range, and the pre-rendering strategy is selected based on the location distribution relationship; Real-time rendering is performed on the models within the adjacent framing range.

[0016] Compared to existing technologies, this invention sets a tag set for each model body, performs viewfinder switching verification on models within a model library, and records several rendering attribute parameters in real time during the viewfinder switching verification process. Based on the temporal variation characteristics of these rendering attribute parameters, it determines the correlation feature values ​​between the tag sets to establish relationships. Subsequently, after model body deployment on the user end, it captures the current viewfinder's framing range and movement direction, predicts adjacent framing ranges based on the movement direction, and renders the models based on the correlation relationships between the current framing range and adjacent models within those ranges. This invention provides adaptive rendering for 3D models in different scene spaces, reducing rendering stutters or anomalies when the viewfinder moves within 3D space, and improving overall image smoothness.

[0017] In particular, this invention sets up tag sets for the model library and performs view switching verification. In reality, due to differences in customer needs and design, the design scenes are highly random, and various 3D models may exist within the scene. Therefore, the viewfinder will capture different 3D models when moving or switching perspectives. During this process, differences in some 3D models may cause rendering stuttering, abrupt changes in image features, and other anomalies. Based on this, this invention considers performing view switching verification in advance. By placing model bodies with different tag sets in a predetermined 3D space, the impact of switching perspectives on rendering when switching model bodies corresponding to different tag sets is determined, and the correlation between tag sets is constructed. Since there are a large number of similar model bodies in the design-type model library, the construction of tag sets can reflect the impact of a type of model body on rendering, which facilitates the rapid subsequent calling of the model body's tag set, rapid analysis of the impact on rendering, and adaptive execution of rendering strategies. This reduces stuttering or rendering anomalies in the rendered images captured when the viewfinder moves in 3D space, and improves image smoothness.

[0018] In particular, this invention predicts the adjacent framing range, and then calls the corresponding tag set of the model objects within the current framing range relative to the adjacent framing range to determine the correlation. Based on this, the model objects are adaptively rendered. In practical applications, the adjacent framing range corresponding to the viewfinder is quickly analyzed. Since the tag set is pre-set, it can be quickly called and its impact on rendering can be analyzed, thereby adaptively executing the rendering strategy. In actual situations, if the model objects with correlation within the adjacent framing range are distributed relatively sparsely, these model objects may cause rendering anomalies after the framing range is switched. Since the dispersion can easily lead to overall image distortion and poor overall visual quality, it is necessary to pre-render all model objects in the image to reduce the abnormality of the rendered image captured by the viewfinder and improve the smoothness of the image. For cases where the distribution is relatively concentrated, the model objects with correlation are relatively concentrated. Therefore, for the relatively concentrated local clustered model objects, all model objects in the area where the local clustered model objects are located are locally pre-rendered, thereby focusing on the areas that are prone to distortion. While saving computing power, this reduces rendering stuttering or anomalies and improves the smoothness of the image. Attached Figure Description

[0019] Figure 1 This is a schematic diagram illustrating the steps of an automated method for generating architectural design scenarios according to an embodiment of the invention. Figure 2 This is a logic block diagram for rendering a model body according to an embodiment of the invention; Figure 3 This is a logic block diagram illustrating the determination of location distribution types in an embodiment of the invention. Figure 4 A logic block diagram for selecting a pre-rendering strategy based on location distribution relationships in an embodiment of the invention. Detailed Implementation

[0020] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0021] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0022] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0023] Please see Figure 1 The diagram illustrates the steps of an automated method for generating architectural design scenes according to an embodiment of the present invention. The automated method for generating architectural design scenes according to an embodiment of the present invention includes: Step S1: Set a tag set for several model objects in the model library. The tag set includes a main tag to identify the model object and several sub-tags that reflect the range of the three-dimensional attributes of the model object. Step S2, performing view switching verification on the model bodies in the model library, includes grouping the model bodies corresponding to different tag sets into a predetermined three-dimensional space, and controlling the viewfinder to switch the view range between the model bodies. Step S3: Record several rendering attribute parameters in real time during the framing switching verification process, and determine the correlation feature values ​​between the tag sets based on the change characteristics of the rendering attribute parameters in the temporal dimension, so as to determine the correlation relationship. Step S4: In response to the user terminal completing the model body deployment, capture the current framing range and movement direction of the viewfinder, and predict the adjacent framing range based on the movement direction; Step S5: Determine the tag set corresponding to the model body within the current framing range relative to the adjacent framing range to determine the association relationship. Render the model body based on the association relationship between the tag sets, including... The location distribution type is determined based on the spatial coordinates of the model objects within the adjacent framing range, and the pre-rendering strategy is selected based on the location distribution relationship; Real-time rendering of the model within the adjacent framing range; The pre-rendering strategy includes global pre-rendering or determining local clustered model bodies, and performing local pre-rendering on all model bodies in the region where the local clustered model body is located.

[0024] Specifically, the types of models in the model library depend on the requirements of the design scenario. For example, the models in the model library that come with the 3D design software can also be self-constructed. For example, common models in architectural design scenarios, such as tables, decorations, sofas, and coffee tables, are constructed by those skilled in the art based on the actual situation, which will not be elaborated further here.

[0025] Specifically, a viewfinder is a virtual camera in three-dimensional space used to collect information within the view frustum. The raw data collected by the virtual camera is rendered and then displayed on a monitor. This is existing technology and will not be elaborated further.

[0026] Specifically, the user can deploy a model in three-dimensional space using existing 3D modeling software to complete the design. After the user deploys the model and completes the scene construction, they can control the viewfinder to move and browse the scene. This is existing technology and will not be elaborated further.

[0027] Specifically, when capturing the current framing range and movement direction of the viewfinder, and predicting the adjacent framing range based on the movement direction, it can simulate the framing range adjacent to the current framing range formed after the viewfinder moves in the current movement direction, thereby realizing the function of predicting the adjacent framing range, which will not be elaborated further.

[0028] The direction of movement is obtained by reading the user's operation instructions, provided that the user has authorized the movement.

[0029] Specifically, the process of setting label sets for several model bodies within the model library includes, Determine the type number corresponding to the type of the model body as the main label; Determine several three-dimensional attributes of the model body, and determine the range of the three-dimensional attributes of the model body corresponding to each of the three-dimensional attributes of the model body; The three-dimensional attribute range of each model body is used as a sub-label of the model body; Each tag set corresponds one-to-one with a model body, and the three-dimensional attributes of the model body are preset.

[0030] Specifically, the type of model refers to the classification of model objects. There is no limitation on the classification method. For example, in architectural design, model objects are divided into table type, coffee table type, home appliance type, etc., which will not be elaborated further.

[0031] In practice, a unique serial number is assigned to each type.

[0032] In practice, the three-dimensional properties of the model include the number of independent materials and the number of triangles. These three-dimensional properties comprehensively reflect the model's impact on rendering.

[0033] In practice, the range of three-dimensional attributes of the model body is predetermined, which is essentially a number of continuous intervals. Among them, the maximum number of independent materials and the maximum number of triangles in a number of model bodies in the model library are predetermined. The number of independent materials corresponds to a relatively small order of magnitude. The number of independent materials from 0 to the maximum number of independent materials is divided into 5 consecutive intervals, thus obtaining the range of independent material numbers for 5 model bodies. Since the number of triangles is on the order of magnitude, we divide the number of triangles from 0 to the maximum number of triangles into 100 consecutive intervals, thus obtaining a range of 100 triangles. This will not be elaborated further.

[0034] Specifically, this involves placing the model bodies in groups within a predetermined three-dimensional space. Two model objects are placed at the same time, and the distance between the model objects must be greater than half of the viewfinder's field of view, and the label sets of the model objects placed at the same time must be different.

[0035] Understandably, the purpose of limiting the distance is to ensure that when switching between views for verification, the viewfinder's field of view can independently capture a single model object, and that the field of view can quickly include another model object when switching.

[0036] Specifically, the process of switching the framing range between the various model bodies by controlling the viewfinder includes, Control the viewfinder to correspond to any model body, so that the initial framing range contains only a single model body; Move the viewfinder until the viewfinder contains only the complete other model body described above; Move the viewfinder again until it corresponds to the initial framing range.

[0037] Specifically, the process of determining the association features between tag sets based on the changes in rendering attribute parameters in the temporal dimension, in order to determine the association relationship, includes: Record rendering attribute parameters, construct temporal variation curves of rendering attribute parameters, determine the amplitude difference ratio between curve segments corresponding to when the view range includes the model body, and determine the amplitude difference ratio as the associated feature value; If the associated feature value is greater than a predetermined associated threshold, it is determined that there is an association between the tag sets. The rendering attribute parameter is either GPU utilization or single-frame rendering time.

[0038] In practice, the calculation process for the amplitude difference ratio includes, Solve for the first average amplitude of the curve segment corresponding to the model body, and then solve for the second average amplitude of the curve segment corresponding to the other model body; The absolute difference between the first average amplitude and the second average amplitude is calculated, the average value between the first average amplitude and the second average amplitude is calculated, and the ratio of the absolute difference to the average value is calculated to obtain the amplitude difference ratio, which reflects the difference in rendering attribute parameters when the viewfinder switches to the two model bodies.

[0039] The correlation threshold is obtained from experiments. Several model bodies are pre-verified by switching the view. Those skilled in the art observe the monitor and filter the records of the view switching verification process when there is screen stuttering or frame drops. The average correlation feature value is determined to reflect the normal situation of the correlation feature value of the model body when the screen is abnormal. In practice, the average correlation feature value is used as the predetermined correlation threshold.

[0040] This invention sets up tag sets for the model library and performs view switching verification. In reality, due to differences in customer needs and design, the design scenes are highly random, and various 3D models may exist within the scene. Therefore, the viewfinder will capture different 3D models when moving or switching perspectives. During this process, differences in some 3D models may cause rendering stuttering, abrupt changes in image features, and other anomalies. Based on this, this invention considers pre-performing view switching verification. By placing model bodies with different tag sets in a predetermined 3D space, the impact of switching perspectives on rendering when switching model bodies corresponding to different tag sets is determined, and the correlation between tag sets is constructed. Since there are a large number of similar model bodies in the design-type model library, the construction of tag sets can reflect the impact of a type of model body on rendering, facilitating the rapid subsequent call to the tag sets of model bodies, quick analysis of the impact on rendering, and adaptive execution of rendering strategies. This reduces stuttering or rendering anomalies in the rendered images captured when the viewfinder moves in 3D space, improving image smoothness.

[0041] Specifically, please refer to Figure 2 As shown, it is a logic block diagram for rendering a model body according to an embodiment of the invention. Rendering the model body based on the association relationship between the corresponding tag sets of model bodies within the current framing range and adjacent framing ranges includes... If there is a correlation between the current field of view and the corresponding tag set of the model body in the adjacent field of view, the position distribution type is determined based on the spatial coordinates of the model body in the adjacent field of view, and the pre-rendering strategy is selected based on the position distribution relationship; If there is no correlation between the label sets of the model objects in the current field of view and those in the adjacent field of view, then the model objects in the adjacent field of view will be rendered in real time.

[0042] This invention predicts the adjacent framing range, then calls the corresponding tag set of the model objects within the current framing range relative to the adjacent framing range to determine the correlation. Based on this, the model objects are adaptively rendered. In practical applications, the adjacent framing range corresponding to the viewfinder is quickly analyzed. Since the tag set is pre-set, it can be quickly called and its impact on rendering can be analyzed, thus adaptively executing the rendering strategy. In actual situations, if the model objects with correlations within the adjacent framing range are scattered, these model objects may cause rendering anomalies after the framing range is switched. Due to the dispersion, the overall image may change, resulting in a poor overall visual experience. Therefore, it is necessary to pre-render all model objects in the image to reduce rendering anomalies captured by the viewfinder and improve image smoothness. For cases where the distribution is relatively concentrated, the model objects with correlations are relatively concentrated. Therefore, local pre-rendering is performed on all model objects within the area where the relatively concentrated clustered model objects are located, thus focusing on areas prone to anomalies. This saves computing power, reduces rendering stuttering or anomalies, and improves image smoothness.

[0043] Specifically, please refer to Figure 3 The diagram shown is a logic block diagram for determining the location distribution type according to an embodiment of the invention. The process of determining the location distribution type based on the spatial coordinates of the model body within the adjacent framing range includes: Determine the spatial coordinates of model bodies within the adjacent framing range that have related label sets; Determine the distances between the spatial coordinates and their nearest neighbor spatial coordinates to solve for the mean distance; If the mean distance is greater than or equal to the distance concentration threshold, it is determined to be a dispersed distribution type; If the mean distance is less than the distance concentration threshold, it is determined to be a concentrated distribution type.

[0044] Specifically, the distance concentration threshold is determined based on the maximum width of the framing range. In practice, considering that when the distance between models is less than 0.2 to 0.3 times the maximum width of the framing range, they are relatively concentrated in terms of line of sight and belong to the field of attention in terms of visual perception, the distance concentration threshold is selected between 0.2 and 0.3 times the maximum width, preferably 0.2 times the maximum width.

[0045] Specifically, please refer to Figure 4 As shown, it is a logic block diagram of selecting a pre-rendering strategy based on location distribution relationship according to an embodiment of the invention. The selection of the pre-rendering strategy based on location distribution relationship includes, For distributed and scattered data types, global pre-rendering is used; For the concentrated and dispersed types, a local cluster model body is determined, and local pre-rendering is performed on all model bodies in the region where the local cluster model body is located.

[0046] Specifically, modern real-time rendering typically renders the model within the viewfinder's field of view in real time, rather than rendering the entire scene. Pre-rendering, on the other hand, renders the model within the corresponding field of view when the viewfinder has not captured the corresponding field of view. This will not be elaborated further.

[0047] Specifically, the process of determining the local cluster model volume and performing local pre-rendering on all model volumes within the region containing the local cluster model volume includes: Based on spatial coordinates, cluster the model bodies that have a relationship with the label set. If the clustering conditions are met, determine the cluster and the model body corresponding to the cluster is determined as the local cluster model body. The field of view is divided into several regions, and all models within the region where the local clustered model is located are pre-rendered locally. The clustering conditions are that the distance between the spatial coordinates of any model body within a cluster is less than a predetermined distance threshold, and the number of model bodies in a cluster must be greater than a predetermined number threshold.

[0048] In practice, the predetermined distance threshold is determined in advance. In order to better reflect the situation where the line of sight is relatively concentrated, the predetermined distance threshold is set to 0.85 times the distance concentration threshold, and the predetermined quantity threshold is set to 2.

[0049] In practice, the framing area can be divided according to a grid pattern. It can be divided into 9 grids, or it can be adjusted to other patterns, which will not be elaborated here.

[0050] The embodiments also provide a system for an automated generation method applied to architectural design scenarios, comprising: The tag segmentation module is used to set tag sets for several model bodies in the model library. The tag set includes a main tag to identify the model body and several sub-tags that reflect the range of the three-dimensional attributes of the model body. The verification module verifies the viewfinder switching of the model objects in the model library, including grouping the model objects corresponding to different tag sets into a predetermined three-dimensional space, and controlling the viewfinder to switch the viewfinder between the model objects. The association determination module is used to record several rendering attribute parameters in real time during the framing switching verification process, and determine the association feature values ​​between the tag sets based on the change characteristics of the rendering attribute parameters in the temporal dimension, so as to determine the association relationship. The range determination module responds to the user terminal to complete the deployment of the model body, captures the current framing range and movement direction of the viewfinder, and predicts the adjacent framing range based on the movement direction. The rendering control module is used to determine the corresponding tag sets of model objects within the current framing range relative to adjacent framing ranges, to determine the association relationship, and to render the model objects based on the association relationship between the tag sets, including... The location distribution type is determined based on the spatial coordinates of the model objects within the adjacent framing range, and the pre-rendering strategy is selected based on the location distribution relationship; Real-time rendering is performed on the models within the adjacent framing range.

[0051] Specifically, there are no restrictions on the specific structure of the tag segmentation module, verification module, association determination module, range determination module, and rendering control module. Each module can be composed of logical components, including field-programmable processors, computers, or microprocessors in computers.

[0052] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. An automated generation method for architectural design scenarios, characterized in that, include: Set a tag set for several model objects in the model library. The tag set includes a main tag to identify the model object and several sub-tags that reflect the range of the three-dimensional attributes of the model object. The framing switching verification for the model objects in the model library includes grouping the model objects corresponding to different tag sets into a predetermined three-dimensional space, and controlling the viewfinder to switch the framing range between the model objects. Several rendering attribute parameters are recorded in real time during the framing switching verification process. Based on the change characteristics of the rendering attribute parameters in the temporal dimension, the correlation feature values ​​between the tag sets are determined to determine the correlation relationship. In response to the completion of model deployment on the user side, the current framing range and movement direction of the viewfinder are captured, and the adjacent framing range is predicted based on the movement direction. Determine the tag sets corresponding to the model objects within the current framing range relative to adjacent framing ranges to establish relationships. Then, render the model objects based on these relationships, including... The location distribution type is determined based on the spatial coordinates of the model objects within the adjacent framing range, and the pre-rendering strategy is selected based on the location distribution relationship; Real-time rendering of the model within the adjacent framing range; The pre-rendering strategy includes global pre-rendering or determining local clustered model bodies, and performing local pre-rendering on all model bodies in the region where the local clustered model body is located.

2. The automated generation method for architectural design scenarios according to claim 1, characterized in that, The process of setting tag sets for several model bodies in the model library includes, Determine the type number corresponding to the type of the model body as the main label; Determine several three-dimensional attributes of the model body, and determine the range of the three-dimensional attributes of the model body corresponding to each of the three-dimensional attributes of the model body; The three-dimensional attribute range of each model body is used as a sub-label of the model body; Each tag set corresponds one-to-one with a model body, and the three-dimensional attributes of the model body are preset.

3. The automated generation method for architectural design scenarios according to claim 1, characterized in that, During the process of grouping the model bodies and placing them in a predetermined three-dimensional space. Two model objects are placed at the same time, and the distance between the model objects must be greater than half of the viewfinder's field of view, and the label sets of the model objects placed at the same time must be different.

4. The automated generation method for architectural design scenarios according to claim 3, characterized in that, The process of switching the framing range between the various model bodies by controlling the viewfinder includes, Control the viewfinder to correspond to any model body, so that the initial framing range contains only a single model body; Move the viewfinder until the viewfinder contains only the complete other model body described above; Move the viewfinder again until it corresponds to the initial framing range.

5. The automated generation method for architectural design scenarios according to claim 4, characterized in that, The process of determining the association features between tag sets based on the changes in rendering attribute parameters in the temporal dimension, in order to establish the association relationship, includes the following steps: Record rendering attribute parameters, construct temporal variation curves of rendering attribute parameters, determine the amplitude difference ratio between curve segments corresponding to when the view range includes the model body, and determine the amplitude difference ratio as the associated feature value; If the associated feature value is greater than a predetermined associated threshold, it is determined that there is an association between the tag sets. The rendering attribute parameter is either GPU utilization or single-frame rendering time.

6. The automated generation method for architectural design scenes according to claim 1, characterized in that, The rendering of the model body is based on the association relationship between the corresponding tag sets of the model body within the current framing range and the adjacent framing range. If there is a correlation between the current field of view and the corresponding tag set of the model body in the adjacent field of view, the position distribution type is determined based on the spatial coordinates of the model body in the adjacent field of view, and the pre-rendering strategy is selected based on the position distribution relationship; If there is no correlation between the label sets of the model objects in the current field of view and those in the adjacent field of view, then the model objects in the adjacent field of view will be rendered in real time.

7. The automated generation method for architectural design scenes according to claim 1, characterized in that, The process of determining the location distribution type based on the spatial coordinates of the model objects within the adjacent framing range includes: Determine the spatial coordinates of model bodies within the adjacent framing range that have related label sets; Determine the distances between the spatial coordinates and their nearest neighbor spatial coordinates to solve for the mean distance; If the mean distance is greater than or equal to the distance concentration threshold, it is determined to be a dispersed distribution type; If the mean distance is less than the distance concentration threshold, it is determined to be a concentrated distribution type.

8. The automated generation method for architectural design scenes according to claim 7, characterized in that, Selecting a pre-rendering strategy based on location distribution relationships includes: For distributed and scattered data types, global pre-rendering is used; For the concentrated and dispersed types, a local cluster model body is determined, and local pre-rendering is performed on all model bodies in the region where the local cluster model body is located.

9. The automated generation method for architectural design scenes according to claim 8, characterized in that, The process of determining the local cluster model volume and performing local pre-rendering of all model volumes within the region containing the local cluster model volume includes: Based on spatial coordinates, cluster the model bodies that have a relationship with the label set. If the clustering conditions are met, determine the cluster and the model body corresponding to the cluster is determined as the local cluster model body. The field of view is divided into several regions, and all models within the region where the local clustered model is located are pre-rendered locally. The clustering conditions are that the distance between the spatial coordinates of any model body within a cluster is less than a predetermined distance threshold, and the number of model bodies in a cluster must be greater than a predetermined number threshold.

10. A system for automatically generating architectural design scenes using the method described in any one of claims 1-9, characterized in that, include: The tag segmentation module is used to set tag sets for several model bodies in the model library. The tag set includes a main tag to identify the model body and several sub-tags that reflect the range of the three-dimensional attributes of the model body. The verification module verifies the viewfinder switching of the model objects in the model library, including grouping the model objects corresponding to different tag sets into a predetermined three-dimensional space, and controlling the viewfinder to switch the viewfinder between the model objects. The association determination module is used to record several rendering attribute parameters in real time during the framing switching verification process, and determine the association feature values ​​between the tag sets based on the change characteristics of the rendering attribute parameters in the temporal dimension, so as to determine the association relationship. The range determination module responds to the user terminal to complete the deployment of the model body, captures the current framing range and movement direction of the viewfinder, and predicts the adjacent framing range based on the movement direction. The rendering control module is used to determine the corresponding tag sets of model objects within the current framing range relative to adjacent framing ranges, to determine the association relationship, and to render the model objects based on the association relationship between the tag sets, including... The location distribution type is determined based on the spatial coordinates of the model objects within the adjacent framing range, and the pre-rendering strategy is selected based on the location distribution relationship; Real-time rendering is performed on the models within the adjacent framing range.

Images