Building design review system based on augmented reality technology

Abstract

The invention discloses a building design review system based on an augmented reality technology. The building design review system comprises a model import and processing module, a scene recognition and positioning module, an augmented reality rendering module, a labeling and commenting module and a data storage and management module. Through an augmented reality (AR) technology, a virtual building model is superposed into a real scene in real time, so that review personnel can feel the real state of a building in an actual environment personally on the scene, and the immersive experience solves the problem that the fusion effect of the building and the surrounding environment cannot be visually presented by traditional two-dimensional drawings and three-dimensional model software; the augmented reality rendering module integrates multi-dimensional data streams such as a building design model, a scene map, annotations and comments into an augmented reality environment to generate a continuous building evolution display sequence, so that the comprehensiveness of review is improved, and the problem that details and spatial feelings are not accurately grasped in a traditional review mode is solved.

Description

Technical Field

[0001] The present invention relates to the field of virtual reality technology, and in particular to an architectural design review system based on augmented reality technology. Background Art

[0002] With the development of the construction industry, architectural designs are becoming increasingly complex, and the requirements for design reviews are becoming increasingly demanding. Traditional review methods rely primarily on paper drawings and computer-generated 3D modeling software. 2D drawings struggle to intuitively represent the three-dimensional spatial relationships of a building, making it difficult for non-professionals to understand. While 3D modeling software can display three-dimensional effects, reviewers can only observe them on a screen, unable to immerse themselves in the building's true state in its actual environment. This limits their ability to assess aspects such as the building's integration with its surroundings and its sense of spatial scale.

[0003] Some architectural design companies use conventional 3D modeling software (such as 3DMAX and SketchUp) to create architectural models, which are then projected onto large screens for presentation and explanation. During the presentation, the designer manually switches between different perspectives and scenes, while reviewers provide comments and suggestions based on the projected images.

[0004] However, when viewing architectural models on a screen or through a projection, reviewers cannot gain an immersive experience and find it difficult to accurately perceive the building's actual spatial size, height, and true relationship with the surrounding environment. They are also not precise enough in their grasp of some details and spatial perceptions.

[0005] In existing technical solutions, the reviewers' interaction methods with the model are relatively simple. They can usually only perform limited rotation and zooming operations through devices such as mice and keyboards. They cannot interact with the model naturally, such as freely moving around in the actual scene to view the building from all angles. This limits the comprehensiveness and depth of the review. Moreover, since the effect of the architectural design in the real scene cannot be intuitively presented, there may be misunderstandings between reviewers and designers when communicating design intentions and modification opinions, resulting in increased communication costs and extended design review cycles. Summary of the Invention

[0006] In order to solve the above-mentioned technical problems, the present invention provides an architectural design review system based on augmented reality technology.

[0007] The technical solution of the present invention is achieved as follows:

[0008] An architectural design review system based on augmented reality technology, comprising a model import and processing module, a scene recognition and positioning module, an augmented reality rendering module, a marking and commenting module, and a data storage and management module;

[0009] The model import and processing module imports architectural design model files in various formats and performs optimization processing to improve rendering efficiency and display effects on AR devices;

[0010] The scene recognition and positioning module collects real-world scene images through the camera of the AR device, uses the SLAM algorithm based on visual feature points to identify key feature points in the scene, and constructs a scene map to achieve accurate positioning and posture tracking of the AR device in the real scene;

[0011] The augmented reality rendering module is based on the AR development platform, which renders the processed building model in real time and superimposes it on the real scene;

[0012] The annotation and comment module provides reviewers with the function of annotating specific parts of the building model through gestures or voice operations, and stores these annotations and comments in association with the corresponding locations on the model;

[0013] The data storage and management module manages various types of data by combining local database and cloud storage.

[0014] Furthermore, the model import and processing module includes a model file selection unit, a model preprocessing unit, an import progress feedback unit and an import result prompt unit;

[0015] The model file selection unit provides an interface button to allow the user to select a building design model file stored locally or in a network;

[0016] The model preprocessing unit optimizes the imported model, including model simplification, material conversion, and texture compression, to improve rendering efficiency on the AR device;

[0017] The import progress feedback unit displays the model import progress in real time and provides pause and cancel operations;

[0018] After the model import is completed, the import result prompt unit displays the import success information and the next step operation guidance.

[0019] Furthermore, the scene recognition and positioning module includes a camera image acquisition unit, a feature point extraction and map construction unit, a positioning and posture tracking unit and an information display unit;

[0020] The camera image acquisition unit acquires real-world scene images captured by the camera of an AR device (such as a smartphone, tablet computer, AR glasses, etc.) in real time, supports camera interfaces of multiple devices, and provides image preprocessing functions;

[0021] The feature point extraction and map construction unit uses a SLAM algorithm based on visual feature points to extract key feature points from the collected images, uses points of different colors to mark the extracted feature points, and dynamically constructs and updates the scene map to ensure the accuracy and completeness of the map;

[0022] The positioning and posture tracking unit calculates the position coordinates and posture angles of the AR device in the real scene in real time based on the feature points and scene map, providing high-precision positioning and posture tracking functions to ensure the accurate placement of the building model in the real scene, support robust tracking in dynamic environments, and reduce tracking losses caused by lighting changes or occlusions;

[0023] The information display unit displays the positioning information of the AR device in real time in the user interface, provides visual feedback, helps users understand the current status of the device, supports map visualization, and displays the construction status and feature point distribution of the current scene map.

[0024] Furthermore, the augmented reality rendering module includes a real-time rendering unit and an environment rendering processing unit;

[0025] The real-time rendering unit renders the optimized building model in real time and overlays it onto the real scene image, supports high frame rate rendering, provides dynamic rendering adjustment function, and optimizes the rendering effect according to device performance and scene complexity;

[0026] The environmental rendering processing unit simulates the real environment, adds realistic environmental effects to the virtual building model, supports dynamic environmental parameter adjustment, updates the virtual model in real time according to environmental changes in the real scene, enhances the material texture of the virtual model, and improves visual realism through texture mapping and surface processing technology.

[0027] Furthermore, the annotation and comment module includes an interaction unit, an annotation tool selection unit, a comment input and attachment adding unit, an annotation and comment storage unit, and a data synchronization unit;

[0028] The interactive unit supports multiple interaction methods, including gesture operations and voice commands, provides an interactive gesture prompt area, displays available gestures and their functional descriptions, supports voice command recognition, and allows users to annotate, comment or operate through voice commands;

[0029] The annotation tool selection unit provides a variety of annotation tools for users to choose from, supports customized annotation styles, and meets different review requirements;

[0030] The comment input and attachment adding unit provides a text input box to support users to enter text comments, and integrates a voice-to-text function to allow users to select pictures from the local album or record voice as comment attachments to enrich the comment content;

[0031] The annotation and comment storage unit associates and stores annotation and comment information with corresponding locations on the building model, supports local and cloud storage, and provides editing, deletion, and modification functions for annotations and comments, making it convenient for users to adjust content;

[0032] The data synchronization unit synchronizes local annotation and comment data to the cloud server in real time, supports sharing and collaboration among team members, and provides offline synchronization function to ensure that the data operated by users can be automatically uploaded in an off-network environment.

[0033] Furthermore, the data storage and management module includes a local database unit, a cloud storage unit, a data synchronization unit and a data security encryption unit;

[0034] The local database unit uses SQLite or other lightweight databases to store data locally on the AR device, storing the model cache data, temporary scene data, and local annotation comment data of the current project, providing fast storage and reading functions and optimizing system operation efficiency;

[0035] The cloud storage unit uses cloud storage services to back up and share data, supports team members to access and collaborate on different devices, and stores complete building design model files, global scene map data, and team-shared annotation and comment data;

[0036] The data synchronization unit performs data upload, download and synchronization operations between the local database and the cloud server to ensure data consistency and integrity, supports automatic synchronization and manual synchronization modes, and provides conflict detection and resolution mechanisms;

[0037] The data security encryption unit encrypts data stored locally and in the cloud to ensure data security and privacy, supports multiple encryption algorithms, and ensures the security of data transmission and storage.

[0038] Furthermore, during the model import process, the model import and processing module extracts the timestamp information of the architectural design model file and aligns it with the system's timeline framework through a preset timestamp alignment algorithm; combined with semantic parsing technology, it automatically generates data format conversion rules based on the natural language description in the architectural design document, and converts heterogeneous model data into a unified format for subsequent processing; based on the update frequency information of the model file, it dynamically adjusts the priority of the model data to ensure that the model parts with high frequency updates can be synchronized to the virtual display in real time.

[0039] Furthermore, during the scene recognition process, the scene recognition and positioning module combines timestamp information to ensure that the construction of the scene map is consistent with the timeline framework of the architectural design model; according to the update frequency of the scene map, the priority of the scene data is dynamically adjusted to ensure real-time synchronization between the scene and the architectural model.

[0040] Furthermore, the augmented reality rendering module integrates multidimensional data streams such as architectural design models, scene maps, annotations and comments into the augmented reality environment according to the timeline framework, generating a continuous architectural evolution display sequence; when the architectural design model or scene map undergoes dynamic adjustment, the incremental update process is triggered to update only the changed parts to ensure the real-time and consistency of the virtual display; the rendered multidimensional data stream is verified through a periodic verification algorithm, and if inconsistent timestamps or data misalignment are found, the repair process is automatically initiated.

[0041] Furthermore, the annotation and comment module associates annotation and comment information with timestamps to ensure their accurate position in the building evolution display sequence, and supports dynamic updates of annotations and comments. When the building design model changes, the incremental update process is automatically triggered to synchronously update the relevant annotation and comment information.

[0042] The data storage and management module stores the timeline framework and related timestamp information in the local database and cloud server to ensure data consistency and integrity; supports the storage and synchronization of multi-dimensional data streams, including architectural design models, scene maps, annotations and comments, and optimizes storage efficiency through an incremental update mechanism; and regularly performs consistency checks on the stored multi-dimensional data streams to ensure data integrity and accuracy.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] 1. This invention uses augmented reality (AR) technology to overlay virtual architectural models onto real-world scenes in real time, allowing reviewers to experience the building's true state in its actual environment. This immersive experience solves the problem that traditional two-dimensional drawings and three-dimensional modeling software cannot intuitively present the integration of the building and its surroundings. The augmented reality rendering module integrates multi-dimensional data streams such as architectural design models, scene maps, annotations, and comments into the augmented reality environment to generate a continuous display sequence of architectural evolution. This not only improves the comprehensiveness of the review process, but also solves the problem of inaccurate grasp of details and spatial perception in traditional review methods.

[0045] 2. The annotation and comment module supports multiple interaction methods, including gestures and voice commands. Reviewers can annotate, comment or operate on the building model through natural gestures or voice commands. This natural interaction method solves the problem of single interaction method in traditional review and improves the convenience and efficiency of review. It provides real-time feedback to confirm that the user's operation has been recognized and executed, and supports dynamic updates of annotations and comments. When the architectural design model changes, the incremental update process is automatically triggered, and the relevant annotation and comment information is updated synchronously to ensure the real-time and consistency of the review process.

[0046] 3. Utilizing the scene recognition and positioning module, the SLAM algorithm enables precise positioning and posture tracking of AR devices in real-world scenarios, ensuring the accurate placement of virtual building models within the real-world scene. This addresses the issue of traditional review methods that prevent accurate perception of the building's actual spatial size, height, and true relationship with the surrounding environment. The system also supports dynamic environmental parameter adjustment, updating the lighting and shadow effects of the virtual model in real time based on lighting changes in the real scene, ensuring a realistic integration of the virtual model with the real environment, further enhancing the accuracy and comprehensiveness of the review.

[0047] 4. Through texture mapping, PBR material models, and real-time path tracing technology, realistic lighting, shadows, and material effects are added to virtual building models, enhancing the immersiveness and visual fidelity of the virtual models. Through visual fusion technology, the virtual model is seamlessly integrated with the real scene, reducing the sense of visual disconnection and enabling reviewers to more intuitively evaluate the effect of architectural design in the actual environment. This addresses the limitations of traditional review methods in evaluating the integration of buildings and surrounding environments.

[0048] 5. The data storage and management module uses a combination of local databases and cloud storage to support the storage and synchronization of multi-dimensional data streams. This not only optimizes data management efficiency but also enables real-time access and collaboration among team members across different devices, resolving the high communication costs and long design cycles associated with traditional reviews. The incremental update mechanism optimizes storage efficiency, not only updating changes and reducing data transmission volume, but also regularly performing consistency checks on stored multi-dimensional data streams to ensure data integrity and accuracy, further enhancing system reliability and collaboration efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a system framework diagram of an architectural design review system based on augmented reality technology according to the present invention;

[0050] Figure 2 This is a framework diagram of the model import and processing module in the present invention;

[0051] Figure 3 This is a framework diagram of the scene recognition and positioning module in the present invention;

[0052] Figure 4 This is a framework diagram of the augmented reality rendering module in the present invention;

[0053] Figure 5 This is a framework diagram of the annotation and comment module in the present invention;

[0054] Figure 6 This is a framework diagram of the data storage and management module in the present invention. DETAILED DESCRIPTION

[0055] In order to make the purposes, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0056] Example 1

[0057] like Figure 1-6 As shown, this embodiment provides an architectural design review system based on augmented reality technology, including a model import and processing module, a scene recognition and positioning module, an augmented reality rendering module, a marking and commenting module, and a data storage and management module;

[0058] The model import and processing module imports architectural design model files in various formats and performs optimization processing to improve rendering efficiency and display effects on AR devices;

[0059] The scene recognition and positioning module collects real-world scene images through the camera of the AR device, uses the SLAM algorithm based on visual feature points to identify key feature points in the scene, and constructs a scene map to achieve accurate positioning and posture tracking of the AR device in the real scene;

[0060] The augmented reality rendering module is based on the AR development platform, which renders the processed building model in real time and superimposes it on the real scene;

[0061] The annotation and comment module provides reviewers with the function of annotating specific parts of the building model through gestures or voice operations, and stores these annotations and comments in association with the corresponding locations on the model;

[0062] The data storage and management module manages various types of data by combining local database and cloud storage.

[0063] Furthermore, the model import and processing module includes a model file selection unit, a model preprocessing unit, an import progress feedback unit and an import result prompt unit;

[0064] The model file selection unit provides an interface button to allow the user to select a building design model file stored locally or on the network; a file selection button is designed through a graphical user interface to support access to the local file system and network path, and a file dialog box is used to implement the file selection function;

[0065] The model preprocessing unit optimizes the imported model, including model simplification, material conversion, and texture compression, to improve rendering efficiency on AR devices; and uses a geometric simplification algorithm to reduce the number of polygons in the model;

[0066] Model simplification rate:

[0067] Convert complex materials in the model into simple materials suitable for AR device rendering, and use texture compression algorithms (such as ETC2 or ASTC) to reduce texture file size;

[0068] The import progress feedback unit displays the model import progress in real time and provides pause and cancel operations; uses a progress bar control to display the import progress, monitors the completion percentage of the import task through a thread, and updates the progress bar in real time;

[0069] After the model import is completed, the import result prompt unit displays the import success information and the next operation guidance; displays the import success information through a pop-up dialog box and provides a prompt for the next operation.

[0070] Furthermore, the scene recognition and positioning module includes a camera image acquisition unit, a feature point extraction and map construction unit, a positioning and posture tracking unit and an information display unit;

[0071] The camera image acquisition unit acquires real-world scene images captured by the camera of an AR device (such as a smartphone, tablet computer, AR glasses, etc.) in real time, supports camera interfaces of multiple devices, and provides image preprocessing functions;

[0072] Preprocessing of the collected images, including denoising, contrast enhancement and color correction, to improve the accuracy of subsequent feature point extraction, including Gaussian filtering (denoising) and histogram equalization (contrast enhancement);

[0073] The feature point extraction and map construction unit uses a SLAM algorithm based on visual feature points to extract key feature points from the collected images, uses points of different colors to mark the extracted feature points, and dynamically constructs and updates the scene map to ensure the accuracy and completeness of the map;

[0074] Use ORB (Oriented FAST and Rotated BRIEF) or SIFT (Scale-Invariant Feature Transform) algorithm to extract key feature points from the image. ORB algorithm combines FAST key point detection and BRIEF description to extract feature points {P1, P2, ..., P n}, feature point descriptor d i For subsequent matching;

[0075] Calculate the 3D point position by matching feature points and triangulating:

[0076] X=Triangulate(P1,P2,d1,d2)

[0077] Use BundleAdjustment to optimize the map's geometry and reduce cumulative errors:

[0078]

[0079] The positioning and posture tracking unit calculates the position coordinates and posture angles of the AR device in the real scene in real time based on the feature points and scene map, providing high-precision positioning and posture tracking functions to ensure the accurate placement of the building model in the real scene, support robust tracking in dynamic environments, and reduce tracking losses caused by lighting changes or occlusions;

[0080] Use the PnP (Perspective-n-Point) algorithm to calculate the AR device's pose. By matching feature points in the image with 3D points in the map, the device's camera pose is estimated.

[0081] The PnP algorithm estimates the pose by minimizing the reprojection error:

[0082]

[0083] Where R and t represent the rotation matrix and translation vector respectively;

[0084] In dynamic environments, multi-view geometry methods (such as ICP, Iterative Closest Point) and deep learning techniques (such as CNN) are used to enhance the robustness of tracking. The ICP algorithm iteratively optimizes the matching error between point clouds:

[0085]

[0086] Where T represents the transformation matrix, X i and Y i Represent the point clouds of the current frame and the reference frame respectively;

[0087] Use Kalman filter or extended Kalman filter (EKF) to smooth the pose estimation to reduce tracking loss caused by lighting changes or occlusion;

[0088] The information display unit displays the positioning information of the AR device in real time in the user interface, provides visual feedback, helps users understand the current status of the device, supports map visualization, and displays the construction status and feature point distribution of the current scene map;

[0089] The user interface displays the AR device's position coordinates (x, y, z) and attitude angles (pitch, yaw, and roll) in real time, and uses a graphical interface (such as a dashboard or digital display) to intuitively display the device's current status.

[0090] Display the construction status of the scene map in real time, including the distribution of feature points and map updates. Use different colors or icons to represent constructed feature points and newly extracted feature points to help users understand the dynamic changes of the map.

[0091] Provide visual feedback, such as prompts when tracking is lost or suggestions for repositioning, and display map integrity and accuracy information in the user interface to help users evaluate the reliability of the system.

[0092] Furthermore, the augmented reality rendering module includes a real-time rendering unit and an environment rendering processing unit;

[0093] The real-time rendering unit renders the optimized building model in real time and overlays it onto the real scene image, supports high frame rate rendering, provides dynamic rendering adjustment function, and optimizes the rendering effect according to device performance and scene complexity;

[0094] Optimize the architectural model before rendering, including model simplification, material conversion and texture compression to improve rendering efficiency;

[0095] Use efficient 3D model formats (such as glTF) to load optimized models to ensure fast loading and rendering;

[0096] Use AR development platforms (such as ARKit, ARCore, or WebXR) to render architectural models in real time and overlay them into real-world scenes;

[0097] Support high frame rate rendering (e.g. 60 frames per second) to ensure smooth interaction between virtual models and real scenes;

[0098] Dynamically adjust rendering parameters based on device performance and scene complexity. For example, reduce rendering accuracy of complex scenes to maintain smoothness.

[0099] Use LOD (level of detail) technology to automatically adjust model details based on the distance between the user and the model;

[0100] Use GPU to accelerate rendering, reduce CPU burden and improve rendering efficiency;

[0101] Generate realistic lighting effects using real-time path tracing techniques such as ray tracing powered by NVIDIA RT Core;

[0102] The environment rendering processing unit simulates the real environment, adds realistic environmental effects to the virtual building model, supports dynamic environmental parameter adjustment, updates the virtual model in real time according to environmental changes in the real scene, enhances the material texture of the virtual model, and improves visual fidelity through texture mapping and surface processing technology;

[0103] Simulate lighting conditions in real environments, add realistic shadows and reflection effects to virtual building models, and use ambient occlusion (AO) technology to enhance the three-dimensionality and realism of the scene;

[0104] Dynamically adjust the lighting parameters of the virtual model according to the lighting changes in the real scene (such as time and weather), use light mapping and shadow mapping technology to optimize the rendering effects of lighting and shadows, and enhance the material texture of the virtual model through texture mapping and surface treatment technology;

[0105] Use PBR (Physically Based Rendering) material model to ensure the authenticity of materials under different lighting conditions;

[0106] Through visual fusion technology, virtual models are seamlessly integrated with real scenes to reduce the sense of visual fragmentation, and mixed reality technology (such as transparency adjustment and reflection effects) is used to enhance the immersion of virtual models.

[0107] Furthermore, the annotation and comment module includes an interaction unit, an annotation tool selection unit, a comment input and attachment adding unit, an annotation and comment storage unit, and a data synchronization unit;

[0108] The interactive unit supports multiple interaction methods, including gesture operations and voice commands, provides an interactive gesture prompt area, displays available gestures and their functional descriptions, supports voice command recognition, and allows users to annotate, comment or operate through voice commands;

[0109] Provides a variety of predefined gesture operations (such as click, slide, zoom, and rotate), allowing users to annotate and operate building models through gestures. Sets an interactive gesture prompt area in the user interface to display available gestures and their functional descriptions to help users quickly get started.

[0110] An integrated speech recognition engine (such as Google Speech-to-Text or Apple Speech Recognizer) supports voice command recognition, allowing users to annotate, comment, or perform actions (such as "add an annotation to the back of the building" or "zoom in on the model") through voice commands;

[0111] Provide real-time feedback to confirm that the user's operation has been recognized and executed by the system. For voice commands, the system can respond with voice or confirm the operation result through interface prompts;

[0112] The annotation tool selection unit provides a variety of annotation tools for users to choose from, supports customized annotation styles, and meets different review requirements;

[0113] Provides a variety of annotation tools, such as text labels, arrows, circles, rectangles, etc. for users to choose from, and supports custom annotation styles, including color, size and transparency adjustment, to meet different review needs;

[0114] A toolbar is set up in the user interface to display all available annotation tools. Users can select the required annotation tool by clicking the icon in the toolbar;

[0115] After selecting the annotation tool, users can use gestures to add annotations at specific locations on the building model. The annotation locations are associated with the 3D coordinates of the building model to ensure the accuracy and consistency of the annotations.

[0116] The comment input and attachment adding unit provides a text input box to support users to enter text comments, and integrates a voice-to-text function to allow users to select pictures from the local album or record voice as comment attachments to enrich the comment content;

[0117] Provides a text input box to support users to enter text comments, and integrates voice-to-text function, allowing users to input comments through voice, and the system automatically converts them into text;

[0118] Allow users to select pictures from local albums or record voice as comment attachments, supporting multiple attachment formats (such as pictures and audio) to enrich comment content;

[0119] Store comments and their attachments in association with the corresponding locations in the building model to ensure the context of the comments is clear;

[0120] The annotation and comment storage unit associates and stores annotation and comment information with corresponding locations on the building model, supports local and cloud storage, and provides editing, deletion, and modification functions for annotations and comments, making it convenient for users to adjust content;

[0121] The annotation and comment information is stored in association with the 3D position of the building model to ensure its accurate position in the building evolution display sequence. Each annotation and comment is accompanied by a timestamp to record its time position in the building evolution process.

[0122] Use a local database (such as SQLite) to store annotation and comment data, supporting fast read and write and local access. Synchronize annotation and comment data to a cloud server (such as AWS S3) to support sharing and collaboration among team members.

[0123] Provides editing, deletion, and modification functions for annotations and comments, making it easy for users to adjust content, supports data version management, and records user operation history for easy traceability and auditing;

[0124] The data synchronization unit synchronizes local annotation and comment data to the cloud server in real time, supports sharing and collaboration among team members, and provides offline synchronization function to ensure that user-operated data can be automatically uploaded in an off-network environment.

[0125] Synchronize locally stored annotation and comment data to the cloud server in real time to ensure data freshness and consistency. It supports automatic synchronization and manual synchronization modes, and users can choose the synchronization method according to their needs.

[0126] In an offline environment, user-operated data will be cached locally. When the device reconnects to the network, the cached data will be automatically uploaded to ensure data integrity and consistency.

[0127] During the synchronization process, data conflicts are detected (e.g., annotations at the same location are modified by different users), and conflict resolution mechanisms such as "last edit takes precedence" or "merge edits" are provided to ensure data accuracy.

[0128] Furthermore, the data storage and management module includes a local database unit, a cloud storage unit, a data synchronization unit and a data security encryption unit;

[0129] The local database unit uses SQLite or other lightweight databases to store data locally on the AR device, storing the model cache data, temporary scene data, and local annotation comment data of the current project, providing fast storage and reading functions and optimizing system operation efficiency;

[0130] The cloud storage unit uses cloud storage services to back up and share data, supports team members to access and collaborate on different devices, and stores complete building design model files, global scene map data, and team-shared annotation and comment data;

[0131] The data synchronization unit performs data upload, download and synchronization operations between the local database and the cloud server to ensure data consistency and integrity, supports automatic synchronization and manual synchronization modes, and provides conflict detection and resolution mechanisms;

[0132] The data security encryption unit encrypts data stored locally and in the cloud to ensure data security and privacy, supports multiple encryption algorithms, and ensures the security of data transmission and storage.

[0133] This embodiment provides an architectural design review system based on augmented reality technology. Its working principle is to achieve the whole process from model import, scene recognition, augmented reality rendering, annotation and commenting to data management through the collaborative work of multiple modules. The specific working process is as follows:

[0134] Model import and processing: The system imports architectural design model files in various formats through the model import and processing module and performs optimization processing, such as model simplification, material conversion, and texture compression, to improve rendering efficiency on AR devices;

[0135] Scene recognition and positioning: The scene recognition and positioning module uses the AR device's camera to capture real-world scene images, extracts key feature points through the SLAM algorithm, and constructs a scene map to achieve precise positioning and posture tracking of the AR device in the real-world scene.

[0136] Augmented Reality Rendering: Based on the AR development platform, the augmented reality rendering module renders the optimized building model in real time and overlays it onto the real scene. It supports high frame rate rendering and dynamic environment parameter adjustment to ensure seamless integration of virtual models and real scenes.

[0137] Annotation and Commenting: The Annotation and Commenting module allows reviewers to annotate and comment on specific parts of the building model through gestures or voice commands, and stores this information in association with the model location. It supports local and cloud storage and data synchronization, facilitating team collaboration.

[0138] Data storage and management: The data storage and management module uses a combination of local databases and cloud storage to manage various types of data, including model caches, scene maps, annotations, and comments, ensuring data consistency and security, and providing encryption processing to protect privacy;

[0139] It has achieved automation and intelligence in the entire process from model import to review, improving the efficiency and accuracy of architectural design review, while supporting real-time collaboration and data sharing among team members on different devices.

[0140] Example 2

[0141] This embodiment provides an architectural design review system based on augmented reality technology, including a model import and processing module, a scene recognition and positioning module, an augmented reality rendering module, a marking and commenting module, and a data storage and management module;

[0142] The model import and processing module imports architectural design model files in various formats and performs optimization processing to improve rendering efficiency and display effects on AR devices;

[0143] The scene recognition and positioning module collects real-world scene images through the camera of the AR device, uses the SLAM algorithm based on visual feature points to identify key feature points in the scene, and constructs a scene map to achieve accurate positioning and posture tracking of the AR device in the real scene;

[0144] The augmented reality rendering module is based on the AR development platform, which renders the processed building model in real time and superimposes it on the real scene;

[0145] The annotation and comment module provides reviewers with the function of annotating specific parts of the building model through gestures or voice operations, and stores these annotations and comments in association with the corresponding locations on the model;

[0146] The data storage and management module manages various types of data by combining local database and cloud storage.

[0147] Furthermore, the model import and processing module includes a model file selection unit, a model preprocessing unit, an import progress feedback unit and an import result prompt unit;

[0148] The model file selection unit provides an interface button to allow the user to select a building design model file stored locally or on the network; a file selection button is designed through a graphical user interface to support access to the local file system and network path, and a file dialog box is used to implement the file selection function;

[0149] The model preprocessing unit optimizes the imported model, including model simplification, material conversion, and texture compression, to improve rendering efficiency on AR devices; and uses a geometric simplification algorithm to reduce the number of polygons in the model;

[0150] Model simplification rate:

[0151] Convert complex materials in the model into simple materials suitable for AR device rendering, and use texture compression algorithms (such as ETC2 or ASTC) to reduce texture file size;

[0152] The import progress feedback unit displays the model import progress in real time and provides pause and cancel operations; uses a progress bar control to display the import progress, monitors the completion percentage of the import task through a thread, and updates the progress bar in real time;

[0153] After the model import is completed, the import result prompt unit displays the import success information and the next operation guidance; displays the import success information through a pop-up dialog box and provides a prompt for the next operation.

[0154] Furthermore, the scene recognition and positioning module includes a camera image acquisition unit, a feature point extraction and map construction unit, a positioning and posture tracking unit and an information display unit;

[0155] The camera image acquisition unit acquires real-world scene images captured by the camera of an AR device (such as a smartphone, tablet computer, AR glasses, etc.) in real time, supports camera interfaces of multiple devices, and provides image preprocessing functions;

[0156] Preprocessing of the collected images, including denoising, contrast enhancement and color correction, to improve the accuracy of subsequent feature point extraction, including Gaussian filtering (denoising) and histogram equalization (contrast enhancement);

[0157] The feature point extraction and map construction unit uses a SLAM algorithm based on visual feature points to extract key feature points from the collected images, uses points of different colors to mark the extracted feature points, and dynamically constructs and updates the scene map to ensure the accuracy and completeness of the map;

[0158] Use ORB (Oriented FAST and Rotated BRIEF) or SIFT (Scale-Invariant Feature Transform) algorithm to extract key feature points from the image. ORB algorithm combines FAST key point detection and BRIEF description to extract feature points {P1, P2, ..., P n}, feature point descriptor d i For subsequent matching;

[0159] Calculate the 3D point position by matching feature points and triangulating:

[0160] X=Triangulate(P1,P2,d1,d2)

[0161] Use BundleAdjustment to optimize the map's geometry and reduce cumulative errors:

[0162]

[0163] The positioning and posture tracking unit calculates the position coordinates and posture angles of the AR device in the real scene in real time based on the feature points and scene map, providing high-precision positioning and posture tracking functions to ensure the accurate placement of the building model in the real scene, support robust tracking in dynamic environments, and reduce tracking losses caused by lighting changes or occlusions;

[0164] Use the PnP (Perspective-n-Point) algorithm to calculate the AR device's pose. By matching feature points in the image with 3D points in the map, the device's camera pose is estimated.

[0165] The PnP algorithm estimates the pose by minimizing the reprojection error:

[0166]

[0167] Where R and t represent the rotation matrix and translation vector respectively;

[0168] In dynamic environments, multi-view geometry methods (such as ICP, Iterative Closest Point) and deep learning techniques (such as CNN) are used to enhance the robustness of tracking. The ICP algorithm iteratively optimizes the matching error between point clouds:

[0169]

[0170] Where T represents the transformation matrix, X i and Y i Represent the point clouds of the current frame and the reference frame respectively;

[0171] Use Kalman filter or extended Kalman filter (EKF) to smooth the pose estimation to reduce tracking loss caused by lighting changes or occlusion;

[0172] The information display unit displays the positioning information of the AR device in real time in the user interface, provides visual feedback, helps users understand the current status of the device, supports map visualization, and displays the construction status and feature point distribution of the current scene map;

[0173] The user interface displays the AR device's position coordinates (x, y, z) and attitude angles (pitch, yaw, and roll) in real time, and uses a graphical interface (such as a dashboard or digital display) to intuitively display the device's current status.

[0174] Display the construction status of the scene map in real time, including the distribution of feature points and map updates. Use different colors or icons to represent constructed feature points and newly extracted feature points to help users understand the dynamic changes of the map.

[0175] Provide visual feedback, such as prompts when tracking is lost or suggestions for repositioning, and display map integrity and accuracy information in the user interface to help users evaluate the reliability of the system.

[0176] Furthermore, the augmented reality rendering module includes a real-time rendering unit and an environment rendering processing unit;

[0177] The real-time rendering unit renders the optimized building model in real time and overlays it onto the real scene image, supports high frame rate rendering, provides dynamic rendering adjustment function, and optimizes the rendering effect according to device performance and scene complexity;

[0178] Optimize the architectural model before rendering, including model simplification, material conversion and texture compression to improve rendering efficiency;

[0179] Use efficient 3D model formats (such as glTF) to load optimized models to ensure fast loading and rendering;

[0180] Use AR development platforms (such as ARKit, ARCore, or WebXR) to render architectural models in real time and overlay them into real-world scenes;

[0181] Support high frame rate rendering (e.g. 60 frames per second) to ensure smooth interaction between virtual models and real scenes;

[0182] Dynamically adjust rendering parameters based on device performance and scene complexity. For example, reduce rendering accuracy of complex scenes to maintain smoothness.

[0183] Use LOD (level of detail) technology to automatically adjust model details based on the distance between the user and the model;

[0184] Use GPU to accelerate rendering, reduce CPU burden and improve rendering efficiency;

[0185] Generate realistic lighting effects using real-time path tracing techniques such as ray tracing powered by NVIDIA RT Core;

[0186] The environment rendering processing unit simulates the real environment, adds realistic environmental effects to the virtual building model, supports dynamic environmental parameter adjustment, updates the virtual model in real time according to environmental changes in the real scene, enhances the material texture of the virtual model, and improves visual fidelity through texture mapping and surface processing technology;

[0187] Simulate lighting conditions in real environments, add realistic shadows and reflection effects to virtual building models, and use ambient occlusion (AO) technology to enhance the three-dimensionality and realism of the scene;

[0188] Dynamically adjust the lighting parameters of the virtual model according to the lighting changes in the real scene (such as time and weather), use light mapping and shadow mapping technology to optimize the rendering effects of lighting and shadows, and enhance the material texture of the virtual model through texture mapping and surface treatment technology;

[0189] Use PBR (Physically Based Rendering) material model to ensure the authenticity of materials under different lighting conditions;

[0190] Through visual fusion technology, virtual models are seamlessly integrated with real scenes to reduce the sense of visual fragmentation, and mixed reality technology (such as transparency adjustment and reflection effects) is used to enhance the immersion of virtual models.

[0191] Furthermore, the annotation and comment module includes an interaction unit, an annotation tool selection unit, a comment input and attachment adding unit, an annotation and comment storage unit, and a data synchronization unit;

[0192] The interactive unit supports multiple interaction methods, including gesture operations and voice commands, provides an interactive gesture prompt area, displays available gestures and their functional descriptions, supports voice command recognition, and allows users to annotate, comment or operate through voice commands;

[0193] Provides a variety of predefined gesture operations (such as click, slide, zoom, and rotate), allowing users to annotate and operate building models through gestures. Sets an interactive gesture prompt area in the user interface to display available gestures and their functional descriptions to help users quickly get started.

[0194] An integrated speech recognition engine (such as Google Speech-to-Text or Apple Speech Recognizer) supports voice command recognition, allowing users to annotate, comment, or perform actions (such as "add an annotation to the back of the building" or "zoom in on the model") through voice commands;

[0195] Provide real-time feedback to confirm that the user's operation has been recognized and executed by the system. For voice commands, the system can respond with voice or confirm the operation result through interface prompts;

[0196] The annotation tool selection unit provides a variety of annotation tools for users to choose from, supports customized annotation styles, and meets different review requirements;

[0197] Provides a variety of annotation tools, such as text labels, arrows, circles, rectangles, etc. for users to choose from, and supports custom annotation styles, including color, size and transparency adjustment, to meet different review needs;

[0198] A toolbar is set up in the user interface to display all available annotation tools. Users can select the required annotation tool by clicking the icon in the toolbar;

[0199] After selecting the annotation tool, users can use gestures to add annotations at specific locations on the building model. The annotation locations are associated with the 3D coordinates of the building model to ensure the accuracy and consistency of the annotations.

[0200] The comment input and attachment adding unit provides a text input box to support users to enter text comments, and integrates a voice-to-text function to allow users to select pictures from the local album or record voice as comment attachments to enrich the comment content;

[0201] Provides a text input box to support users to enter text comments, and integrates voice-to-text function, allowing users to input comments through voice, and the system automatically converts them into text;

[0202] Allow users to select pictures from local albums or record voice as comment attachments, supporting multiple attachment formats (such as pictures and audio) to enrich comment content;

[0203] Store comments and their attachments in association with the corresponding locations in the building model to ensure the context of the comments is clear;

[0204] The annotation and comment storage unit associates and stores annotation and comment information with corresponding locations on the building model, supports local and cloud storage, and provides editing, deletion, and modification functions for annotations and comments, making it convenient for users to adjust content;

[0205] The annotation and comment information is stored in association with the 3D position of the building model to ensure its accurate position in the building evolution display sequence. Each annotation and comment is accompanied by a timestamp to record its time position in the building evolution process.

[0206] Use a local database (such as SQLite) to store annotation and comment data, supporting fast read and write and local access. Synchronize annotation and comment data to a cloud server (such as AWS S3) to support sharing and collaboration among team members.

[0207] Provides editing, deletion, and modification functions for annotations and comments, making it easy for users to adjust content, supports data version management, and records user operation history for easy traceability and auditing;

[0208] The data synchronization unit synchronizes local annotation and comment data to the cloud server in real time, supports sharing and collaboration among team members, and provides offline synchronization function to ensure that user-operated data can be automatically uploaded in an off-network environment.

[0209] Synchronize locally stored annotation and comment data to the cloud server in real time to ensure data freshness and consistency. It supports automatic synchronization and manual synchronization modes, and users can choose the synchronization method according to their needs.

[0210] In an offline environment, user-operated data will be cached locally. When the device reconnects to the network, the cached data will be automatically uploaded to ensure data integrity and consistency.

[0211] During the synchronization process, data conflicts are detected (e.g., annotations at the same location are modified by different users), and conflict resolution mechanisms such as "last edit takes precedence" or "merge edits" are provided to ensure data accuracy.

[0212] Furthermore, the data storage and management module includes a local database unit, a cloud storage unit, a data synchronization unit and a data security encryption unit;

[0213] The local database unit uses SQLite or other lightweight databases to store data locally on the AR device, storing the model cache data, temporary scene data, and local annotation comment data of the current project, providing fast storage and reading functions and optimizing system operation efficiency;

[0214] The cloud storage unit uses cloud storage services to back up and share data, supports team members to access and collaborate on different devices, and stores complete building design model files, global scene map data, and team-shared annotation and comment data;

[0215] The data synchronization unit performs data upload, download and synchronization operations between the local database and the cloud server to ensure data consistency and integrity, supports automatic synchronization and manual synchronization modes, and provides conflict detection and resolution mechanisms;

[0216] The data security encryption unit encrypts data stored locally and in the cloud to ensure data security and privacy, supports multiple encryption algorithms, and ensures the security of data transmission and storage.

[0217] This embodiment provides an architectural design review system based on augmented reality technology. Its working principle is to achieve the whole process from model import, scene recognition, augmented reality rendering, annotation and commenting to data management through the collaborative work of multiple modules. The specific working process is as follows:

[0218] Model import and processing: The system imports architectural design model files in various formats through the model import and processing module and performs optimization processing, such as model simplification, material conversion, and texture compression, to improve rendering efficiency on AR devices;

[0219] Scene recognition and positioning: The scene recognition and positioning module uses the AR device's camera to capture real-world scene images, extracts key feature points through the SLAM algorithm, and constructs a scene map to achieve precise positioning and posture tracking of the AR device in the real-world scene.

[0220] Augmented Reality Rendering: Based on the AR development platform, the augmented reality rendering module renders the optimized building model in real time and overlays it onto the real scene. It supports high frame rate rendering and dynamic environment parameter adjustment to ensure seamless integration of virtual models and real scenes.

[0221] Annotation and Commenting: The Annotation and Commenting module allows reviewers to annotate and comment on specific parts of the building model through gestures or voice commands, and stores this information in association with the model location. It supports local and cloud storage and data synchronization, facilitating team collaboration.

[0222] Data storage and management: The data storage and management module uses a combination of local databases and cloud storage to manage various types of data, including model caches, scene maps, annotations, and comments, ensuring data consistency and security, and providing encryption processing to protect privacy;

[0223] It has achieved automation and intelligence in the entire process from model import to review, improving the efficiency and accuracy of architectural design review, while supporting real-time collaboration and data sharing among team members on different devices.

[0224] Furthermore, during the model import process, the model import and processing module extracts the timestamp information of the architectural design model file and aligns it with the system's timeline framework using a preset timestamp alignment algorithm. Combining semantic parsing technology, the module automatically generates data format conversion rules based on the natural language description in the architectural design document, converting heterogeneous model data into a unified format for subsequent processing. Based on the update frequency information of the model file, the module dynamically adjusts the priority of the model data to ensure that the model parts with high frequency updates can be synchronized to the virtual display in real time.

[0225] Specifically, the process includes the following:

[0226] When importing the model, the timestamp information of different stages (conceptual design stage, detailed design stage, construction stage) is extracted from the architectural design model file. These timestamp information reflects the key time nodes of the model in each design stage;

[0227] A multi-step reasoning framework is used to preliminarily process the extracted timestamps to generate preliminary time series data. Deep learning techniques (such as LSTM or Transformer) are used to perform feature learning on the preliminary time series data to obtain temporal feature information.

[0228] Based on the time feature information, a custom module is generated to intelligently extract and align timestamps, rearrange the processed timestamp data in chronological order, generate a unified timeline framework, and align the timestamps of the model file with the system's timeline framework;

[0229] Obtain natural language descriptions in architectural design documents, parse the semantic information using natural language processing technology, and extract descriptions and requirements about data formats from the documents through semantic parsing;

[0230] Based on the semantic information obtained from the analysis, the mapping relationships between different data formats (such as Revit, AutoCAD, SketchUp, etc.) are derived, and a data format conversion rule library is pre-established to store these mapping relationships;

[0231] Introducing the custom module generation function, allowing users to describe data conversion rules in natural language, parsing the user's natural language description, and automatically generating the corresponding conversion module;

[0232] Through the format conversion engine, heterogeneous model data (such as sketch models, structural analysis data, progress information, etc.) are converted into a unified system format for subsequent processing;

[0233] Obtain the update frequency information of the model file and monitor the update frequency of the model at different stages;

[0234] The preset threshold is used to determine the update rate of each stage. Based on the comparison between the update rate and the preset value, the priority of the model data is determined to dynamically adjust the demand;

[0235] For models that are updated frequently, a dynamic scheduling algorithm is used to optimize the transmission order of data streams to ensure that these data can be synchronized to the virtual display system first;

[0236] During the synchronization process, the update rate changes of multi-dimensional data are continuously monitored, and the scheduling strategy is dynamically adjusted according to real-time data to ensure efficient response of the system.

[0237] Furthermore, the scene recognition and positioning module combines timestamp information during the scene recognition process to ensure that the construction of the scene map is consistent with the timeline framework of the building design model; dynamically adjusts the priority of the scene data according to the update frequency of the scene map to ensure real-time synchronization between the scene and the building model;

[0238] Specifically, the process includes the following:

[0239] At the beginning of scene recognition, the timestamp information of the current stage is extracted from the timeline framework of the architectural design model, covering the conceptual design, detailed design and construction stages. These timestamp information reflects the key time nodes of the model in each design stage;

[0240] Use a SLAM algorithm based on visual feature points (such as ORB-SLAM) to extract key feature points in the scene in real time and dynamically build a scene map. During the feature point extraction process, combine the timestamp information and assign a timestamp label to each feature point to ensure that it is consistent with the timeline framework of the architectural design model;

[0241] Correct the timestamps of feature points in the scene map using a preset timestamp alignment algorithm to align them with the time axis of the architectural design model. This includes: comparing the timestamps of feature points in the scene map with the timestamps of the architectural design model; adjusting the timestamps of feature points to align them with the time axis of the architectural design model, ensuring the consistency of the scene map and the architectural design model in the time dimension.

[0242] Monitor the update frequency of the scene map in real time and calculate the update rate of feature points. The update frequency reflects the dynamic changes of the scene map at different stages;

[0243] Based on the preset update rate threshold, the priority adjustment needs of the scene map are determined. If the update frequency of the scene map is higher than the preset threshold, its priority is increased to ensure that the scene data with high frequency updates can be synchronized to the virtual display system first;

[0244] For high-priority scene data, a dynamic scheduling algorithm is used to optimize the transmission order of data streams. Specifically, this includes: prioritizing the processing and transmission of frequently updated feature point data, continuously monitoring the update rate changes of multi-dimensional data, and dynamically adjusting the scheduling strategy based on real-time data to ensure efficient system response;

[0245] During the scene map update process, the timeline consistency between the scene data and the architectural design model is continuously monitored. If timestamp inconsistency or data misalignment is detected, the incremental update mechanism is triggered to repair and update only the affected data.

[0246] Adopting an incremental update mechanism to repair and update the affected scene data, specifically including: determining the scope of the affected data, repairing and updating the affected data to ensure that it is consistent with the timeline of the building design model;

[0247] The repaired data is realigned to the timeline framework to ensure the consistency of the scene map and the architectural design model in the time dimension;

[0248] In the scene recognition process, a multi-step reasoning framework and deep learning technology (such as LSTM or Transformer) are introduced to perform feature learning on the time series data of feature points, thereby enhancing the stability of feature points in dynamic environments.

[0249] For lighting changes, occlusions, or dynamic scene changes, deep learning models are used to optimize feature points to ensure the accuracy and robustness of scene maps. Specifically, deep learning models are used to optimize feature points to reduce tracking loss due to environmental changes; and the weights of feature points are dynamically adjusted to ensure the accuracy and real-time performance of high-priority feature points.

[0250] Furthermore, the augmented reality rendering module integrates multidimensional data streams such as architectural design models, scene maps, annotations, and comments into the augmented reality environment based on a timeline framework, generating a continuous sequence of architectural evolution displays. When the architectural design model or scene map undergoes dynamic adjustments, an incremental update process is triggered to update only the changed parts, ensuring the real-time and consistency of the virtual display. A periodic verification algorithm is used to verify the rendered multidimensional data streams, and if inconsistent timestamps or data misalignment are found, a repair process is automatically initiated.

[0251] Specifically, the process includes the following:

[0252] Obtain multi-dimensional data streams such as architectural design models, scene maps, annotations, and comments from the timeline framework, and integrate these data streams according to the unified format of the timeline framework to ensure their consistency in the time dimension;

[0253] Based on the data interface of the augmented reality environment, the integrated multi-dimensional data stream is input into the augmented reality environment, and a continuous architectural evolution display sequence is generated through real-time rendering technology, allowing users to intuitively observe the evolution of architectural design in the augmented reality environment;

[0254] Real-time monitoring of dynamic adjustments to architectural design models or scene maps. When changes are detected, the stage attributes of the changed data are identified and their position on the timeline is determined.

[0255] If the changed data belongs to a certain stage, the incremental update process is triggered and the incremental update algorithm is used to extract only the newly added or modified parts of the changed data, avoiding repeated processing of the unchanged parts;

[0256] Determine the mapping position of the changed data on the timeline according to the logical order of the timeline. Use the data synchronization algorithm to match the changed data with the existing data in the virtual display to ensure data consistency.

[0257] If there are conflicts in the matching results (such as inconsistent timestamps or data misalignment), a conflict resolution mechanism is used to adjust the data to ensure that the changed data can be correctly integrated into the virtual display;

[0258] Obtain the status information of the rendered multi-dimensional data stream and use the timeline framework to monitor the data stream in real time;

[0259] According to the preset periodic rules, the consistency verification algorithm is triggered to verify the data stream. The verification content includes timestamp consistency, data integrity, and data continuity on the time axis.

[0260] If timestamp inconsistency or data misalignment is detected, the affected data range is determined and an incremental update mechanism is used to repair and update only the affected data, avoiding reprocessing of the entire data stream.

[0261] The repaired data is realigned into the timeline framework to ensure the overall structural consistency of the data stream. Through the correction algorithm, the overall structure of the data stream is adjusted to restore its correct display in the augmented reality environment.

[0262] Furthermore, the annotation and comment module associates annotation and comment information with timestamps to ensure their accurate position in the building evolution display sequence, and supports dynamic updates of annotations and comments. When the building design model changes, the incremental update process is automatically triggered to synchronously update the relevant annotation and comment information.

[0263] Specifically, the process includes the following:

[0264] When annotations and comments are generated, the current timestamp is extracted from the timeline framework of the architectural design model and associated with the annotation or comment information. Each annotation and comment is accompanied by a timestamp to record its specific time position in the architectural evolution process.

[0265] Bind annotations and comments to specific locations on the architectural design model to ensure their accurate placement in the architectural evolution display sequence. Use spatial positioning technology to precisely place annotations and comments on the corresponding parts of the architectural model.

[0266] Support users to edit, delete or modify existing annotations and comments, and update the content of annotations and comments in real time while maintaining their association with timestamps;

[0267] When users modify annotations and comments, the system automatically records the updated content and marks it as "updated". The updated annotations and comments are realigned with the building evolution display sequence based on the timestamp to ensure their accuracy in the time dimension.

[0268] Monitor changes in architectural design models in real time, including new, modified, or deleted model parts. When a model change is detected, identify the phase attributes of the changed data and determine its position on the timeline.

[0269] If the changed data belongs to a specific stage, the incremental update process is automatically triggered, extracting only the annotations and comments related to the changed data, avoiding repeated processing of unaffected parts;

[0270] According to the logical order of the timeline, the changed annotations and comments are matched with the existing data in the virtual display. The data synchronization algorithm ensures that the annotations and comments can accurately reflect the latest status of the architectural design model.

[0271] If timestamp conflicts or data misalignment are found during the synchronization process, a conflict resolution mechanism will be used to adjust the data to ensure the accuracy and consistency of annotation and comment information.

[0272] The data storage and management module stores the timeline framework and related timestamp information in the local database and cloud server to ensure data consistency and integrity; supports the storage and synchronization of multi-dimensional data streams, including architectural design models, scene maps, annotations and comments, and optimizes storage efficiency through an incremental update mechanism; and regularly performs consistency checks on the stored multi-dimensional data streams to ensure data integrity and accuracy.

[0273] Specifically, the process includes the following:

[0274] Store the timeline frame and related timestamp information in a local database (such as SQLite) to ensure fast data reading and writing and local access efficiency. The local database is used to store the current project's model cache data, temporary scene data, and local annotation and comment data.

[0275] Synchronize the timeline framework and related timestamp information to a cloud server (such as AWS S3 or other cloud storage services) to ensure data backup and sharing among team members. The cloud storage is used to store the complete building design model file, global scene map data, and team-shared annotation and comment data.

[0276] Multi-dimensional data streams, including architectural design models, scene maps, annotations, and comments, are stored in local databases and cloud servers respectively. Each data stream is timestamped to ensure its traceability over time.

[0277] When a part of a multidimensional data stream changes, only the changed part is updated instead of re-uploading the entire data stream. This incremental update mechanism optimizes storage efficiency and reduces data transmission volume.

[0278] Supports automatic synchronization and manual synchronization modes. In automatic synchronization mode, the system triggers synchronization operations based on preset time intervals or data changes; in manual synchronization mode, users can manually trigger data synchronization to ensure consistency between local and cloud data.

[0279] Regularly perform consistency checks on stored multi-dimensional data streams to ensure data integrity and accuracy. Verification includes timestamp consistency, data integrity, and data continuity along the timeline.

[0280] Encrypt data stored locally and in the cloud to ensure data security and privacy. Support multiple encryption algorithms to ensure the security of data transmission and storage.

[0281] Set strict access permissions to ensure that only authorized users can access and operate stored data, and prevent data leakage and unauthorized access through authentication and authorization mechanisms.

[0282] While the specific embodiments of the present invention have been described in detail above, they are intended only as examples, and the present invention is not limited to the specific embodiments described above. Those skilled in the art should understand that the above embodiments and descriptions are merely illustrative of the principles of the present invention, and that various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and modifications are intended to fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. An architectural design review system based on augmented reality technology, characterized by: It includes model import and processing module, scene recognition and positioning module, augmented reality rendering module, annotation and comment module and data storage and management module; The model import and processing module imports architectural design model files in various formats and performs optimization processing to improve rendering efficiency and display effects on AR devices; The scene recognition and positioning module collects real-world scene images through the camera of the AR device, uses the SLAM algorithm based on visual feature points to identify key feature points in the scene, and constructs a scene map to achieve accurate positioning and posture tracking of the AR device in the real scene; The augmented reality rendering module is based on the AR development platform, which renders the processed building model in real time and superimposes it on the real scene; The annotation and comment module provides reviewers with the function of annotating specific parts of the building model through gestures or voice operations, and stores these annotations and comments in association with the corresponding locations on the model; The data storage and management module manages various types of data by combining local database and cloud storage.

2. The architectural design review system based on augmented reality technology according to claim 1, characterized in that: The model import and processing module includes a model file selection unit, a model preprocessing unit, an import progress feedback unit and an import result prompt unit; The model file selection unit provides an interface button to allow the user to select a building design model file stored locally or in a network; The model preprocessing unit optimizes the imported model, including model simplification, material conversion, and texture compression, to improve rendering efficiency on the AR device; The import progress feedback unit displays the model import progress in real time and provides pause and cancel operations; After the model import is completed, the import result prompt unit displays the import success information and the next step operation guidance.

3. The architectural design review system based on augmented reality technology according to claim 1, characterized in that: The scene recognition and positioning module includes a camera image acquisition unit, a feature point extraction and map construction unit, a positioning and posture tracking unit and an information display unit; The camera image acquisition unit acquires real-world scene images captured by the camera of an AR device (such as a smartphone, tablet computer, AR glasses, etc.) in real time, supports camera interfaces of multiple devices, and provides image preprocessing functions; The feature point extraction and map construction unit uses a SLAM algorithm based on visual feature points to extract key feature points from the collected images, uses points of different colors to mark the extracted feature points, and dynamically constructs and updates the scene map to ensure the accuracy and completeness of the map; The positioning and posture tracking unit calculates the position coordinates and posture angles of the AR device in the real scene in real time based on the feature points and scene map, providing high-precision positioning and posture tracking functions to ensure the accurate placement of the building model in the real scene, support robust tracking in dynamic environments, and reduce tracking losses caused by lighting changes or occlusions; The information display unit displays the positioning information of the AR device in real time in the user interface, provides visual feedback, helps users understand the current status of the device, supports map visualization, and displays the construction status and feature point distribution of the current scene map.

4. The architectural design review system based on augmented reality technology according to claim 1, characterized in that: The augmented reality rendering module includes a real-time rendering unit and an environment rendering processing unit; The real-time rendering unit renders the optimized building model in real time and overlays it onto the real scene image, supports high frame rate rendering, provides dynamic rendering adjustment function, and optimizes the rendering effect according to device performance and scene complexity; The environmental rendering processing unit simulates the real environment, adds realistic environmental effects to the virtual building model, supports dynamic environmental parameter adjustment, updates the virtual model in real time according to environmental changes in the real scene, enhances the material texture of the virtual model, and improves visual realism through texture mapping and surface processing technology.

5. The architectural design review system based on augmented reality technology according to claim 1, characterized in that: The annotation and comment module includes an interaction unit, an annotation tool selection unit, a comment input and attachment adding unit, an annotation and comment storage unit and a data synchronization unit; The interactive unit supports multiple interaction methods, including gesture operations and voice commands, provides an interactive gesture prompt area, displays available gestures and their functional descriptions, supports voice command recognition, and allows users to annotate, comment or operate through voice commands; The annotation tool selection unit provides a variety of annotation tools for users to choose from, supports customized annotation styles, and meets different review requirements; The comment input and attachment adding unit provides a text input box to support users to enter text comments, and integrates a voice-to-text function to allow users to select pictures from the local album or record voice as comment attachments to enrich the comment content; The annotation and comment storage unit associates and stores annotation and comment information with corresponding locations on the building model, supports local and cloud storage, and provides editing, deletion, and modification functions for annotations and comments, making it convenient for users to adjust content; The data synchronization unit synchronizes local annotation and comment data to the cloud server in real time, supports sharing and collaboration among team members, and provides offline synchronization function to ensure that the data operated by users can be automatically uploaded in an off-network environment.

6. The architectural design review system based on augmented reality technology according to claim 5, characterized in that: The data storage and management module includes a local database unit, a cloud storage unit, a data synchronization unit and a data security encryption unit; The local database unit uses SQLite or other lightweight databases to store data locally on the AR device, storing the model cache data, temporary scene data, and local annotation comment data of the current project, providing fast storage and reading functions and optimizing system operation efficiency; The cloud storage unit uses cloud storage services to back up and share data, supports team members to access and collaborate on different devices, and stores complete building design model files, global scene map data, and team-shared annotation and comment data; The data synchronization unit performs data upload, download and synchronization operations between the local database and the cloud server to ensure data consistency and integrity, supports automatic synchronization and manual synchronization modes, and provides conflict detection and resolution mechanisms; The data security encryption unit encrypts data stored locally and in the cloud to ensure data security and privacy, supports multiple encryption algorithms, and ensures the security of data transmission and storage.

7. The architectural design review system based on augmented reality technology according to claim 2, characterized in that: During the model import process, the model import and processing module extracts the timestamp information of the architectural design model file and aligns it with the system's timeline framework using a preset timestamp alignment algorithm; Combined with semantic parsing technology, data format conversion rules are automatically generated based on the natural language description in the architectural design documents, converting heterogeneous model data into a unified format for subsequent processing; based on the update frequency information of the model file, the priority of the model data is dynamically adjusted to ensure that the model parts with high frequency updates can be synchronized to the virtual display in real time.

8. The architectural design review system based on augmented reality technology according to claim 3, characterized in that: During the scene recognition process, the scene recognition and positioning module combines timestamp information to ensure that the construction of the scene map is consistent with the timeline framework of the architectural design model; according to the update frequency of the scene map, the priority of the scene data is dynamically adjusted to ensure real-time synchronization between the scene and the architectural model.

9. The architectural design review system based on augmented reality technology according to claim 4, characterized in that: The augmented reality rendering module integrates multi-dimensional data streams such as architectural design models, scene maps, annotations and comments into the augmented reality environment according to the timeline framework to generate a continuous architectural evolution display sequence; When the architectural design model or scene map undergoes dynamic adjustments, the incremental update process is triggered to update only the changed parts, ensuring the real-time and consistency of the virtual display; The rendered multi-dimensional data stream is verified through a periodic verification algorithm. If inconsistent timestamps or data misalignment are found, the repair process is automatically started.

10. The architectural design review system based on augmented reality technology according to claim 6, characterized in that: The annotation and comment module associates annotation and comment information with timestamps to ensure their accurate position in the architectural evolution display sequence. It also supports dynamic updates of annotations and comments. When the architectural design model changes, it automatically triggers the incremental update process and synchronously updates the relevant annotation and comment information. The data storage and management module stores the timeline frame and related timestamp information in the local database and cloud server to ensure the consistency and integrity of the data; It supports the storage and synchronization of multi-dimensional data streams, including architectural design models, scene maps, annotations and comments, and optimizes storage efficiency through an incremental update mechanism; it regularly performs consistency checks on stored multi-dimensional data streams to ensure data integrity and accuracy.

Images