Sfm mapping scale recovery method and system based on terminal real scale tracking

Abstract

The application relates to a method and system for scale recovery of SFM mapping based on terminal real scale tracking, wherein the method comprises the following steps: acquiring image data, and constructing a three-dimensional model without real scale through an SFM algorithm according to the image data; calculating real scale tracks in the image data through a random consistent similarity transformation algorithm based on a camera model to obtain scale coefficients of the tracks and inlier points; sorting the scale coefficients of the tracks from small to large, calculating a quartile range, filtering abnormal scale coefficients through the quartile range, obtaining effective scale coefficients, and obtaining a globally optimal scale coefficient through weighted average of the effective scale coefficients according to the inlier points; and performing scale recovery through the optimal scale coefficient to obtain a three-dimensional model with real scale. Through the application, the reliability of scale recovery of SFM mapping is improved.

Description

Technical Field

[0001] This application relates to the field of augmented reality technology, and in particular to a method and system for SFM mapping scale restoration based on terminal real-scale tracking. Background Technology

[0002] Scale recovery is one of the main problems in monocular visual odometry. To solve this problem, the methods used in academia and industry to calculate the scale of monocular visual odometry and SFM (struct from motion) scale are mainly divided into two types. One type requires the fusion of other sensors, such as lidar or inertial sensors; the other type utilizes known scales in the environment, such as placing calibration plates or measuring the size of feature objects.

[0003] Specifically, in some published patent documents, the solution disclosed in patent publication number CN113256804A uses sensors, such as GPS, to obtain the geographical location of trajectory nodes and convert it to scale data in a reference coordinate system as the first trajectory data. Then, scale-free visual 3D reconstruction is performed on multiple first trajectory data, and scale recovery is performed based on the reconstructed position corresponding to the first trajectory. The trajectory data of this method comes entirely from the sensor results, lacking filtering for possible outliers, which leads to inaccurate scale recovery. The solution disclosed in patent publication number CN113779012A uses the height of a monocular camera to the ground for scale recovery, and performs average filtering on the noise data during this process. However, this filtering method lacks effective statistical reference, affecting the accuracy and consistency of the scale.

[0004] Currently, no effective solution has been proposed to address the issue of low reliability in 3D mapping scale recovery in related technologies. Summary of the Invention

[0005] This application provides a method and system for SFM mapping scale recovery based on terminal real-scale tracking, in order to at least solve the problem of low reliability of 3D mapping scale recovery in related technologies.

[0006] In a first aspect, embodiments of this application provide a method for SFM mapping scale recovery based on terminal real-scale tracking, the method comprising:

[0007] Image data is acquired, and a three-dimensional model without real scale is constructed based on the image data using the SFM algorithm.

[0008] The true-scale trajectory in the image data is calculated using a camera-model-based random consistency similarity transformation algorithm to obtain the trajectory's scale coefficient and number of inliers.

[0009] The scale coefficients of the trajectory are sorted from smallest to largest, the interquartile range is calculated, and abnormal scale coefficients are filtered out by the interquartile range to obtain the effective scale coefficients. The effective scale coefficients are then weighted and averaged by the number of interior points to obtain the globally optimal scale coefficients. The optimal scale coefficients are then used to restore the scale and obtain a three-dimensional model with the true scale.

[0010] In some embodiments, acquiring image data includes:

[0011] An ordered set of image sequences and their corresponding poses in the trajectory coordinate system are acquired through the interactive terminal module. The acquired data is used for real-scale restoration.

[0012] An ordered set of image sequences is obtained by extracting frames from video, and an unordered set of images is obtained by shooting. The acquired data is used for SFM mapping.

[0013] In some embodiments, the acquisition of an ordered set of image sequences and their corresponding poses in the trajectory coordinate system via the interactive terminal module includes:

[0014] An ordered set of image sequences and their poses in the corresponding trajectory coordinate system are collected based on a preset threshold for real-scale restoration.

[0015] In some embodiments, the method includes, prior to calculating the true-scale trajectory in the image data using a camera-model-based stochastic consistency similarity transform algorithm:

[0016] The images corresponding to the real-scale trajectories are registered in the SFM 3D model map.

[0017] In some embodiments, the true-scale trajectory in the image data is calculated using a camera-model-based stochastic consistency similarity transformation algorithm, and the resulting trajectory's scale coefficient and number of inliers include:

[0018] Obtain the pose set of the real-scale trajectory image sequence in the trajectory coordinate system and the pose set in the coordinate system of the 3D model without real scale;

[0019] By transforming the pose set in the 3D model coordinate system without real scale to the trajectory coordinate system through similarity transformation, the scale coefficients are obtained. At the same time, abnormal images and abnormal poses are filtered out through random consistency algorithm.

[0020] In some embodiments, the abnormal image and abnormal pose include:

[0021] Abnormal images or poses are generated when the image sequence corresponding to the true-scale trajectory is registered incorrectly.

[0022] In some embodiments, the interactive terminal module includes: a mobile communication device and an AR device.

[0023] Secondly, embodiments of this application provide a system for SFM mapping scale restoration based on terminal real-scale tracking, the system comprising:

[0024] The SFM mapping module is used to acquire image data and construct a three-dimensional model without real scale based on the image data using the SFM algorithm.

[0025] The scale recovery module is used to calculate the true scale trajectory in the image data using a camera-model-based stochastic consistency similarity transformation algorithm, to obtain the trajectory's scale coefficients and the number of inliers.

[0026] The scale coefficients of the trajectory are sorted from smallest to largest, the interquartile range is calculated, and abnormal scale coefficients are filtered out by the interquartile range to obtain the effective scale coefficients. The effective scale coefficients are then weighted and averaged by the number of interior points to obtain the globally optimal scale coefficients. The optimal scale coefficients are then used to restore the scale and obtain a three-dimensional model with the true scale.

[0027] Thirdly, embodiments of this application provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the SFM mapping scale restoration method based on terminal real-scale tracking as described in the first aspect above.

[0028] Fourthly, embodiments of this application provide a storage medium storing a computer program that, when executed by a processor, implements the SFM mapping scale restoration method based on terminal real-scale tracking as described in the first aspect above.

[0029] Compared to related technologies, the SFM mapping scale restoration method based on terminal real-scale tracking provided in this application acquires image data and constructs a three-dimensional model without real scale based on the image data using the SFM algorithm; calculates the real-scale trajectory in the image data using a random consistency similarity transformation algorithm based on the camera model to obtain the trajectory's scale coefficient and number of inliers; sorts the trajectory's scale coefficients from smallest to largest, calculates the interquartile range, filters out abnormal scale coefficients using the interquartile range to obtain effective scale coefficients, and performs a weighted average of the effective scale coefficients using the number of inliers to obtain the globally optimal scale coefficient; and performs scale restoration using the optimal scale coefficient to obtain a three-dimensional model with real scale.

[0030] This application uses multiple real-scale trajectories, each treated as an independent measurement result, and performs random consistency similarity transformation algorithm calculations separately to obtain multiple scale coefficients. This approach can simultaneously address local image registration errors in the SFM 3D reconstruction model and local scale collapse or expansion of the real-scale trajectory, reducing anomalous images and poses. Furthermore, this application performs interquartile range statistics on the multiple scale coefficients to filter out anomalous scale coefficients, and obtains a more reliable globally optimal scale coefficient through weighted averaging, thereby performing scale recovery and obtaining a real-scale mapping model. Applying this application to augmented reality scenarios solves the problem of low reliability in 3D mapping scale recovery in related technologies and improves the reliability of SFM mapping scale recovery. Attached Figure Description

[0031] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0032] Figure 1 This is a flowchart of a method for SFM mapping scale restoration based on terminal real-scale tracking according to an embodiment of this application;

[0033] Figure 2 This is a structural block diagram of a system for SFM mapping scale recovery based on terminal real-scale tracking according to an embodiment of this application;

[0034] Figure 3 This is a schematic diagram of the internal structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of this application clearer, the application is described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the application. All other embodiments obtained by those skilled in the art based on the embodiments provided in this application without inventive effort are within the scope of protection of this application. Furthermore, it is understood that although the efforts made in such a development process may be complex and lengthy, for those skilled in the art related to the content disclosed in this application, modifications to design, manufacturing, or production based on the technical content disclosed in this application are merely conventional technical means and should not be construed as insufficient disclosure of the content of this application.

[0036] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that is mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0037] Unless otherwise defined, the technical or scientific terms used in this application shall have the ordinary meaning understood by one of ordinary skill in the art to which this application pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this application do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to these processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” used in this application means two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can represent: A alone, A and B simultaneously, and B alone. The terms “first,” “second,” “third,” etc., used in this application are merely to distinguish similar objects and do not represent a specific ordering of the objects.

[0038] This embodiment provides a method for SFM mapping scale restoration based on terminal real-scale tracking. Figure 1 This is a flowchart of a method for SFM mapping scale recovery based on terminal real-scale tracking according to an embodiment of this application, as shown below. Figure 1 As shown, the specific implementation process of the solution includes the following:

[0039] Step S1: Acquire image data and construct a three-dimensional model without real scale based on the image data using the SFM algorithm;

[0040] Preferred, such as Figure 1As shown, in this embodiment, data A for SFM mapping and data B for real-scale restoration are acquired respectively. Data A includes an ordered set of image sequences obtained through video frame extraction, such as {image sequence 1, ..., image sequence N}, and also includes an unordered set of images obtained through photography. Data B is an ordered set of image sequences and their corresponding poses in a trajectory coordinate system acquired through an interactive terminal module. Specifically, this ordered set of image sequences and their corresponding poses in a trajectory coordinate system can be acquired through spatial computing modules on mobile communication devices such as mobile phones, such as ARKit, ARCore, and Huawei AREngine. Alternatively, it can also be acquired through spatial computing modules on AR devices, such as AR glasses, typically binocular VIO, etc., to acquire the ordered set of image sequences and their corresponding poses in a trajectory coordinate system.

[0041] Furthermore, based on the data A obtained above for SFM mapping, a three-dimensional model without real scale is constructed using the SFM algorithm.

[0042] Preferably, when collecting data B for real-scale reconstruction, it is necessary to collect data according to a preset threshold. Specifically, in this embodiment, multiple trajectory data for real-scale reconstruction are collected, and the number of these data is greater than or equal to the preset threshold, thereby ensuring sufficient data volume to guarantee the statistical validity of the data in subsequent calculations. In addition, sufficient real trajectory data can also cross-check whether there are inconsistencies between different real-scale trajectories, and check whether there are scale inconsistencies in the SFM map.

[0043] Preferably, in this embodiment, the preset threshold X is set to 7 by default.

[0044] Step S2: Calculate the true scale trajectory in the image data using a random consistency similarity transformation algorithm based on the camera model to obtain the trajectory's scale coefficient and number of inliers;

[0045] Preferably, before calculating the true-scale trajectory in the image data using the Random Consistent Similarity Transformation Algorithm (RANSAC-sim3) based on the camera model, this embodiment obtains true-scale trajectories, such as ARKit for iOS, ARCore for Android on mobile devices, and binocular VIO on AR glasses, and registers the images corresponding to these trajectories into the SFM 3D model map, i.e., the 3D model without true scale, as a scale reference.

[0046] Furthermore, for each real-scale trajectory in data B, a random consistency similarity transformation algorithm based on the camera model is calculated individually to obtain the scale coefficient and the number of interior points for each trajectory.

[0047] Specifically, since a trajectory is a set of poses composed of multiple temporal sequences, for each trajectory and all the image sequences used to recover the true scale, we can obtain the pose set in the trajectory coordinate system (referred to as scale_traj) and the pose set in the three-dimensional model coordinate system without the true scale (referred to as noscale_traj).

[0048] Then, through similarity transformation, the pose set noscale_traj in the coordinate system of the 3D model without real scale is transformed to the coordinate system where scale_traj is located, and the scale coefficients can be obtained. At the same time, on the basis of the similarity transformation algorithm, the random consistency algorithm can be added to filter out abnormal images and abnormal poses.

[0049] It should be noted that abnormal images and abnormal poses include: abnormal images or abnormal poses generated when the image sequence corresponding to the true-scale trajectory is registered incorrectly.

[0050] This embodiment can simultaneously address situations where local image registration errors occur in the SFM 3D reconstruction model and local scale collapse or expansion occurs in the true-scale trajectory, reducing abnormal images and poses and improving the accuracy and reliability of subsequent scale recovery.

[0051] Step S3: Sort the scale coefficients of the trajectory from smallest to largest, calculate the interquartile range, filter out abnormal scale coefficients through the interquartile range to obtain the effective scale coefficients, and calculate the weighted average of the effective scale coefficients by the number of interior points to obtain the globally optimal scale coefficients. Scale recovery is performed using the optimal scale coefficients to obtain a three-dimensional model with the true scale.

[0052] In this embodiment, all scale coefficients obtained in step S2 are sorted from smallest to largest, their interquartile ranges are calculated, and the abnormal scale coefficients are filtered out by the interquartile ranges to obtain the effective scale coefficients.

[0053] Next, the effective scaling coefficients are weighted and averaged according to the number of inliers to obtain the globally optimal scaling coefficients. The larger the number of inliers, the higher the weight, and the more reliable the result.

[0054] Finally, scale recovery is performed using the globally optimal scale coefficients to obtain a true-scale 3D model.

[0055] In this embodiment, the interquartile range needle can simulate a Gaussian distribution without any statistical prior, analyzing the local scale collapse (underestimation) or expansion (overestimation) of the trajectory itself, i.e., analyzing the scale consistency of SFM mapping. Furthermore, this embodiment also uses a weighted averaging method to obtain a globally optimal scale coefficient, making the final result more reliable.

[0056] Through steps S1 to S3 above, this embodiment uses multiple real-scale trajectories, each as an independent measurement result, and performs random consistency similarity transformation algorithm calculations separately to obtain multiple scale coefficients. This can simultaneously address local image registration errors in the SFM 3D reconstruction model and local scale collapse or expansion of the real-scale trajectory, reducing abnormal images and poses. Furthermore, this application performs interquartile range statistics on the multiple scale coefficients to filter out abnormal scale coefficients, and obtains a more reliable globally optimal scale coefficient through weighted averaging, thereby performing scale recovery and obtaining a real-scale mapping model. Applying this application to augmented reality scenarios solves the problem of low reliability in 3D mapping scale recovery in related technologies and improves the reliability of SFM mapping scale recovery.

[0057] It should be noted that the steps shown in the above process or in the flowchart of the accompanying figures can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0058] This embodiment also provides a system for SFM mapping scale restoration based on terminal real-scale tracking. This system is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the terms "module," "unit," "subunit," etc., can refer to a combination of software and / or hardware that performs a predetermined function. Although the apparatus described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0059] Figure 2 This is a structural block diagram of a system for SFM mapping scale recovery based on terminal real-scale tracking according to an embodiment of this application, as shown below. Figure 2 As shown, the system includes an SFM mapping module 21 and a scale restoration module 22:

[0060] The SFM mapping module 21 is used to acquire image data and construct a 3D model without real scale based on the image data using the SFM algorithm. The scale recovery module 22 is used to calculate the real scale trajectory in the image data using the random consistency similarity transformation algorithm based on the camera model, obtain the scale coefficient and the number of inliers of the trajectory, sort the scale coefficients of the trajectory from smallest to largest, calculate the interquartile range, filter out abnormal scale coefficients by the interquartile range, obtain the effective scale coefficient, and perform a weighted average of the effective scale coefficients by the number of inliers to obtain the globally optimal scale coefficient. The optimal scale coefficient is used to perform scale recovery to obtain a 3D model with real scale.

[0061] Through the aforementioned system, this embodiment uses multiple real-scale trajectories, each treated as an independent measurement result, and performs random consistency similarity transformation algorithm calculations separately to obtain multiple scale coefficients. This approach can simultaneously address local image registration errors in the SFM 3D reconstruction model and local scale collapse or expansion of the real-scale trajectory, reducing abnormal images and poses. Furthermore, this application performs interquartile range statistics on the multiple scale coefficients to filter out abnormal scale coefficients, and obtains a more reliable globally optimal scale coefficient through weighted averaging, thereby performing scale recovery and obtaining a real-scale mapping model. Applying this application to augmented reality scenarios solves the problem of low reliability in 3D mapping scale recovery in related technologies and improves the reliability of SFM mapping scale recovery.

[0062] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated here.

[0063] Furthermore, it should be noted that the aforementioned modules can be either functional modules or program modules, and can be implemented through software or hardware. For modules implemented in hardware, these modules can reside in the same processor; alternatively, they can be located in different processors in any combination.

[0064] This embodiment also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0065] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0066] Furthermore, in conjunction with the SFM mapping scale restoration method based on terminal real-scale tracking in the above embodiments, this application embodiment can provide a storage medium for implementation. This storage medium stores a computer program; when executed by a processor, the computer program implements any of the SFM mapping scale restoration methods based on terminal real-scale tracking in the above embodiments.

[0067] In one embodiment, a computer device is provided, which may be a terminal. The computer device includes a processor, memory, a network interface, a display screen, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements a method for SFM mapping scale recovery based on terminal real-scale tracking. The display screen may be a liquid crystal display (LCD) or an e-ink display. The input device may be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0068] In one embodiment, Figure 3 This is a schematic diagram of the internal structure of an electronic device according to an embodiment of this application, such as... Figure 3 As shown, an electronic device is provided, which can be a server, and its internal structure diagram can be as follows. Figure 3 As shown, the electronic device includes a processor, a network interface, internal memory, and non-volatile memory connected via an internal bus. The non-volatile memory stores the operating system, computer programs, and a database. The processor provides computing and control capabilities, the network interface communicates with external terminals via a network connection, the internal memory provides an environment for the operation of the operating system and computer programs, and the computer programs, when executed by the processor, implement a method for SFM mapping scale recovery based on terminal real-scale tracking. The database stores data.

[0069] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the electronic device to which the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0070] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0071] Those skilled in the art should understand that the technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0072] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for scale restoration in SFM mapping based on terminal real-scale tracking, characterized in that, The method includes: Image data is acquired, and a three-dimensional model without real scale is constructed based on the image data using the SFM algorithm. The scale coefficient and number of inliers of the trajectory are obtained by calculating multiple real-scale trajectories in the image data using a random consistency similarity transformation algorithm based on the camera model. The scale coefficients of the trajectory are sorted from smallest to largest, the interquartile range is calculated, and abnormal scale coefficients are filtered out by the interquartile range to obtain the effective scale coefficients. The effective scale coefficients are then weighted and averaged by the number of interior points to obtain the globally optimal scale coefficients. The optimal scale coefficients are then used to restore the scale and obtain a three-dimensional model with the true scale.

2. The method according to claim 1, characterized in that, The acquisition of image data includes: An ordered set of image sequences and their corresponding poses in the trajectory coordinate system are acquired through the interactive terminal module. The acquired data is used for real-scale restoration. An ordered set of image sequences is obtained by extracting frames from video, and an unordered set of images is obtained by shooting. The acquired data is used for SFM mapping.

3. The method according to claim 2, characterized in that, The ordered set of image sequences and their corresponding poses in the trajectory coordinate system acquired through the interactive terminal module include: An ordered set of image sequences and their poses in the corresponding trajectory coordinate system are collected based on a preset threshold for real-scale restoration.

4. The method according to claim 1, characterized in that, Before calculating the true-scale trajectory in the image data using a camera-model-based stochastic consistency similarity transformation algorithm, the method includes: The images corresponding to the real-scale trajectories are registered in the SFM 3D model map.

5. The method according to any one of claims 1-4, characterized in that, The true-scale trajectory in the image data is calculated using a camera-model-based stochastic consistency similarity transformation algorithm, yielding the trajectory's scale coefficient and number of inliers, including: Obtain the pose set of the real-scale trajectory image sequence in the trajectory coordinate system and the pose set in the coordinate system of the 3D model without real scale; By transforming the pose set in the 3D model coordinate system without real scale to the trajectory coordinate system through similarity transformation, the scale coefficients are obtained. At the same time, abnormal images and abnormal poses are filtered out through random consistency algorithm.

6. The method according to claim 5, characterized in that, The abnormal images and abnormal poses include: Abnormal images or poses are generated when the image sequence corresponding to the true-scale trajectory is registered incorrectly.

7. The method according to claim 2, characterized in that, The interactive terminal module includes: mobile communication devices and AR devices.

8. A system for SFM mapping scale restoration based on terminal real-scale tracking, characterized in that, The system includes: The SFM mapping module is used to acquire image data and construct a three-dimensional model without real scale based on the image data using the SFM algorithm. The scale restoration module is used to calculate multiple true-scale trajectories in the image data using a camera-model-based stochastic consistency similarity transformation algorithm, to obtain the scale coefficients and inliers of the trajectories. The scale coefficients of the trajectory are sorted from smallest to largest, the interquartile range is calculated, and abnormal scale coefficients are filtered out by the interquartile range to obtain the effective scale coefficients. The effective scale coefficients are then weighted and averaged by the number of interior points to obtain the globally optimal scale coefficients. The optimal scale coefficients are then used to restore the scale and obtain a three-dimensional model with the true scale.

9. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the SFM mapping scale restoration method based on terminal real-scale tracking as described in any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute, at runtime, the method for SFM mapping scale restoration based on terminal real-scale tracking as described in any one of claims 1 to 7.

Images