A three-dimensional static model reconstruction method and device

Abstract

The embodiments of the present application disclose a three-dimensional static model reconstruction method and device. The method includes collecting multiple picture sequences of a target object, and performing binary segmentation on each picture sequence to respectively obtain corresponding contour maps; obtaining a three-dimensional visualization shell structure based on the contour maps; obtaining a number of three-dimensional model point cloud data within each contour map in the three-dimensional visualization shell structure based on a preset color consistency calculation criterion; and reconstructing the three-dimensional model point cloud data. The technical solution provided by the embodiments of the present application collects different picture sequences from multiple perspectives for the target object, performs binary segmentation on the picture sequences, then obtains a three-dimensional visualization shell structure, and combines the color consistency information to perform three-dimensional reconstruction on the three-dimensional visualization shell structure. The algorithm of the present application has certain superiority and robustness, can effectively solve the material problem of three-dimensional model printing in additive manufacturing, and provides strong support for obtaining three-dimensional models at low cost.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of image processing technology, and in particular to a three-dimensional static model reconstruction method and device. Background Art

[0002] The technical challenges that need to be solved in the collection and modeling of three-dimensional scenes are mainly convenience, low cost, and real-time issues, which are also the technical points that this patent needs to solve. In combination with the additive manufacturing technology system that plays a vital role in the national economic and social development, it is necessary to model the additive manufacturing smart space in the field of complex industrial manufacturing. The modeling method of multi-dimensional media perception can track three-dimensional scenes statically or dynamically, and restore the real scene in real time to obtain a model with high authenticity and confidence. This reconstruction requires low cost, high convenience, and real-time characteristics. For additive manufacturing technology, it models physical entities through an information platform, and combined with multi-dimensional perception theory, it can ensure that the additive manufacturing model is accurate and realistic. It ensures the accuracy and reliability of low-cost reconstruction models of products for additive manufacturing, and ensures the convenience, extensiveness and timeliness of 3D printing. In particular, the large-scale application of multi-dimensional media sensing equipment to establish product models in the field of intelligent manufacturing has considerable economic benefits. Compared with laser scanners, the advantages of using multi-dimensional media sensing equipment are not only widespread, but also the models provided in 3D printing are highly real-time, highly realistic, and low-cost. They can be widely used in heavy and light industrial fields such as aviation manufacturing, automobile production, machine tool processing, construction, e-commerce, product 3D customization, toy models, etc. The reconstruction process based on multi-dimensional media perception can provide a rich source of materials for 3D printing. The materials are not only convenient to source, but also involve all aspects of production and life, reducing the cost of manufacturing. Summary of the invention

[0003] The embodiments of the present application provide a three-dimensional static simulation reconstruction method, device, equipment and storage medium to reduce manufacturing costs.

[0004] In a first aspect, an embodiment of the present application provides a three-dimensional static model reconstruction method, comprising:

[0005] Collect multiple image sequences of the target object, and perform binary segmentation on each image sequence to obtain the corresponding contour map;

[0006] Obtaining a three-dimensional visualized shell structure based on the contour map;

[0007] Acquire a plurality of three-dimensional model point cloud data in each contour image of the visualized shell structure based on a preset color consistency calculation criterion;

[0008] The three-dimensional model point cloud data is reconstructed.

[0009] Further, obtaining a three-dimensional visualizable housing structure based on the contour map includes:

[0010] Setting visual geometric information based on the camera calibration principle, and obtaining the light cone region of each contour map according to the visual geometric information;

[0011] Calculating the spatial intersection of all light cone regions, and defining the spatial intersection as the visualizable housing structure.

[0012] Further, obtaining a number of three-dimensional model point cloud data within each contour map of the visualizable housing structure based on a preset color consistency calculation criterion includes:

[0013] Selecting any one contour map as the current target contour map, and calculating the color consistency between the current target contour map and adjacent contour maps;

[0014] Selecting spatial three-dimensional point coordinates that meet preset requirements as candidate three-dimensional model point clouds for the current target contour map;

[0015] Selecting a number of three-dimensional model point clouds from the candidate three-dimensional model point clouds as three-dimensional model point cloud data.

[0016] Further, the formula for calculating the color consistency between the current target contour map and adjacent contour maps is:

[0017]

[0018] where NCC is the color consistency; n represents the spatial intersection of n light cone regions, j represents the sequence value of pixels in a 3×3 rectangular window; p is the target pixel in the target view; s is the index number of the candidate depth; n j is the pixel in a 3×3 region block in the target view; f s (n j ) is the pixel corresponding to n j in the current target contour map, and represent the average color intensities of two regions respectively.

[0019] Further, the spatial three-dimensional point coordinates that meet the preset requirements are obtained through the following formula:

[0020]

[0021] where s * is the spatial three-dimensional coordinate point.

[0022] Further, before reconstructing the three-dimensional model point cloud data, it further includes:

[0023] Downsample the 3D model point cloud data.

[0024] Further, the downsampling of the 3D model point cloud data includes:

[0025] Select any two adjacent contour maps and calculate the intersection points of the two adjacent contour maps;

[0026] Extract the hidden points in the stereoscopic visualization shell structure;

[0027] Retain the intersection points and hidden points.

[0028] In a second aspect, an embodiment of the present application provides a 3D static model reconstruction device, including:

[0029] Picture acquisition and segmentation module: used to acquire multiple picture sequences of the target object and perform binary segmentation on each picture sequence to respectively obtain corresponding contour maps;

[0030] Picture stereoscopic visualization module: used to obtain a stereoscopic visualization shell structure based on the contour maps;

[0031] Point cloud data acquisition module: used to acquire a number of 3D model point cloud data within each contour map of the visualization shell structure based on a preset color consistency calculation criterion;

[0032] 3D model reconstruction module: used to reconstruct the 3D model point cloud data.

[0033] Further, the obtaining of the stereoscopic visualization shell structure based on the contour maps includes:

[0034] Set visual geometric information based on the camera calibration principle, and obtain the light cone region of each contour map according to the visual geometric information;

[0035] Calculate the spatial intersection of all light cone regions, and define the spatial intersection as the visualization shell structure.

[0036] Further, the acquisition of a number of 3D model point cloud data within each contour map of the visualization shell structure based on a preset color consistency calculation criterion includes:

[0037] Select any one contour map as the current target contour map, and calculate the color consistency between the current target contour map and the adjacent contour maps;

[0038] Select the spatial three-dimensional point coordinates that meet the preset requirements as the candidate 3D model points of the current target contour map;

[0039] Select a number of 3D model points from the candidate 3D model points as the 3D model point cloud data.

[0040] Further, the formula for calculating the color consistency between the current target contour map and the adjacent contour maps is as follows:

[0041]

[0042] where NCC is the color consistency; n represents the spatial intersection of n light cone regions, j represents the sequence value of pixels in a 3×3 rectangular window; p is the target pixel in the target view; s is the index number of the candidate depth; n j is the pixel in a 3×3 region block in the target view; f s (n j ) is the pixel corresponding to n j in the current target contour map, and represent the average color intensities of the two regions respectively.

[0043] Further, the three-dimensional spatial point coordinates meeting the preset requirements are obtained through the following formula:

[0044]

[0045] where s * is the three-dimensional spatial coordinate point.

[0046] Further, before reconstructing the three-dimensional model point cloud data, the following steps are also included:

[0047] Downsample the three-dimensional model point cloud data.

[0048] Further, the downsampling of the three-dimensional model point cloud data includes:

[0049] Select any two adjacent contour maps and calculate the intersection points of the two adjacent contour maps;

[0050] Extract the hidden points in the stereoscopic visualization shell structure;

[0051] Retain the intersection points and the hidden points.

[0052] In a third aspect, an embodiment of the present application provides a computer device, including: a memory and one or more processors;

[0053] The memory is used to store one or more programs;

[0054] When the one or more programs are executed by the one or more processors, the one or more processors implement a three-dimensional static model reconstruction method as described in the first aspect.

[0055] In a fourth aspect, an embodiment of the present application provides a storage medium containing computer-executable instructions, which are used to execute a three-dimensional static model reconstruction method as described in the first aspect when executed by a computer processor.

[0056] In the embodiment of the present application, different picture sequences are collected from multiple perspectives for the target object, binary segmentation is performed on the picture sequences, and then a three-dimensional visual hull structure is obtained. Combining the color consistency information, three-dimensional reconstruction of the three-dimensional visual hull structure is carried out. The algorithm of the present application has certain superiority and robustness, can effectively solve the material problem of three-dimensional model printing in additive manufacturing, and provides strong support for obtaining three-dimensional models at low cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 is a flowchart of a three-dimensional static model reconstruction method provided by an embodiment of the present application;

[0058] Figure 2 is a schematic structural diagram of a three-dimensional static model reconstruction device provided by an embodiment of the present application;

[0059] Figure 3 is a schematic structural diagram of a computer device provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] In order to make the objectives, technical solutions, and advantages of the present application clearer, the following further describes the specific embodiments of the present application in detail with reference to the drawings. It can be understood that the specific embodiments described herein are only used to explain the present application, rather than limiting the present application. Additionally, it should be noted that for the sake of description, only parts related to the present application are shown in the drawings rather than all the content. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the operations (or steps) as sequential processes, many of the operations can be performed in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. When the operations are completed, the process can be terminated, but there can also be additional steps not included in the drawings. The process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0061] The embodiments of the present application provide a three-dimensional static model reconstruction method and apparatus. In the embodiments of the present application, different picture sequences are collected from multiple perspectives of a target object, binary segmentation is performed on the picture sequences, and then a three-dimensional visual hull structure is obtained. Combining the color consistency information, three-dimensional reconstruction of the three-dimensional visual hull structure is performed. The algorithm of the present application has certain superiority and robustness, can effectively solve the material problem of three-dimensional model printing in additive manufacturing, and provides strong support for obtaining three-dimensional models at low cost.

[0062] The following will be described in detail respectively.

[0063] Figure 1 The flowchart provided by the embodiments of the present application is given. The three-dimensional static model reconstruction method provided by the embodiments of the present application can be executed by a three-dimensional static model reconstruction apparatus. The three-dimensional static model reconstruction apparatus can be implemented in a hardware and / or software manner and integrated in a computer device.

[0064] The following describes the three-dimensional static model reconstruction method executed by the three-dimensional static model reconstruction apparatus as an example. Refer to Figure 1 , the three-dimensional static model reconstruction method includes:

[0065] 101: Collect multiple picture sequences of a target object, and perform binary segmentation on each picture sequence to obtain corresponding contour maps respectively.

[0066] In the embodiment, the three-dimensional model is reconstructed based on the pictures of the target object. Accordingly, first select the target object, and collect picture sequences from different angles of the target object respectively. It is easy to understand that in the embodiment, a camera can be used to collect at least one picture sequence for each vision of the target object. Since the target object is usually a three-dimensional figure in the application of the embodiment, when using a camera to collect picture sequences, corresponding to each face of the three-dimensional figure, collections are made respectively. In order to collect more comprehensively, usually several picture sequences are collected for each face, or each perspective respectively. After collecting enough picture sequences, binary segmentation is performed on each picture sequence. In a picture, there are a target object, a background, and noise. If you want to directly extract the target object from a multi-valued digital image, a common method is to set a threshold T, and use T to divide the data of the image into two parts, the pixel group greater than T and the pixel group less than T, that is, the binaryzation of the image. That is to say, a relatively reasonable threshold is selected to determine whether each pixel point in the image should belong to the target area or the background area, so as to generate the corresponding binary image. The contour map obtained through branch segmentation is also the target area obtained through image binaryzation.

[0067] 102: Obtain a three-dimensional visual hull structure based on the contour map.

[0068] In the embodiment, a picture sequence of a target object is collected by the above camera. Specifically, in the embodiment, visual geometric information is set based on the camera calibration principle, and the light cone region of each contour map is obtained according to the visual geometric information; the spatial intersection of all the light cone regions is calculated, and the spatial intersection is defined as the three-dimensional visualization shell structure.

[0069] In the above, camera calibration includes traditional camera calibration method, active vision camera calibration method, camera self-calibration method, and zero-distortion camera calibration method. In the process of image measurement and machine vision applications, in order to determine the mutual relationship between the three-dimensional geometric position of a certain point on the surface of a spatial object and its corresponding point in the image, it is necessary to establish a geometric model of camera imaging, and these geometric model parameters are camera parameters. Under most conditions, these parameters must be obtained through experiments and calculations, and this process of solving the parameters is called camera calibration (or camera calibration). Whether in image measurement or machine vision applications, the calibration of camera parameters is a very crucial link, and the accuracy of its calibration results and the stability of the algorithm directly affect the accuracy of the results generated by the camera's work. The embodiment can select any one of the existing mature camera calibration methods for parameter calibration, so as to set the visual geometric information. Preferably, the CVH algorithm is used in the embodiment to obtain the three-dimensional visualization shell structure.

[0070] 103: Obtain a number of three-dimensional model point cloud data in each contour map in the three-dimensional visualization shell structure based on a preset color consistency calculation criterion.

[0071] In the embodiment, the three-dimensional model point cloud data is obtained through the obtained three-dimensional visualization shell structure for the reconstruction of the three-dimensional model. In order to make the finally reconstructed three-dimensional model more conform to the target object, the embodiment considers the color consistency.

[0072] Specifically, select any one contour map as the current target contour map, calculate the color consistency between the current target contour map and the adjacent contour map; select the spatial three-dimensional point coordinates that meet the preset requirements as the candidate three-dimensional model point cloud of the current target contour map; select a number of three-dimensional model point clouds from the candidate three-dimensional model point clouds as the three-dimensional model point cloud data.

[0073] In the above, the formula for calculating the color consistency between the current target contour map and the adjacent contour map is:

[0074]

[0075] Among them, NCC is the color consistency; n represents that there are n spatial intersections of light cone regions, j represents the sequence value of pixels in a 3×3 rectangular window; p is the target pixel in the target view; s is the index number of the candidate depth; n j is the pixel in the region block with a size of 3×3 in the target view; fs (n j ) is the pixel corresponding to n in the current target contour map. j And and respectively represent the average color intensities of two regions.

[0076] The spatial three-dimensional point coordinates that meet the preset requirements are obtained through the following formula:

[0077]

[0078] where s * is the spatial three-dimensional coordinate point.

[0079] Finding its color consistency usually refers to the color consistency between the contour maps adjacent to the left and right of the current target contour map. Assuming the number of candidate depths is m, the fear three-dimensional coordinates obtained by generally satisfying the above formula can be used as viewpoints to obtain partial point clouds.

[0080] 104: Reconstruct the three-dimensional model point cloud data.

[0081] In this step, that is, fuse the partial point clouds obtained in the previous step. Specifically, before reconstructing the three-dimensional model point cloud data, downsampling the three-dimensional model point cloud data is also included. Select any two adjacent contour maps, calculate the intersection points of the two adjacent contour maps; extract the hidden points in the stereo visualization shell structure; retain the intersection points and hidden points.

[0082] The new point cloud synthesized from the point clouds of multiple viewpoints contains a large amount of redundant information and incorrect points. To reduce the amount of data for subsequent processing, downsample the point cloud data, and consider the confidence of the data during the downsampling process. In the embodiment, retain the edge points and hidden points, and filter out other points. Edge points refer to the points that exist on the stereo visualization shell structure and must belong to the object surface. Edge points are obtained by the intersection operator of the contour lines of two contour maps, and their normal lines are perpendicular to the contour section. Whether a point belongs to an edge point can be judged by back-projecting each image contour point into the three-dimensional space to form a ray and judging the number of intersection points of the ray and the stereo visualization shell structure. Hidden points are the set of points on the stereo visualization shell structure that cannot be observed by any camera. When extracting hidden points, back-project the points on the stereo visualization shell structure onto each contour image, and judge whether the point is a hidden point by observing whether the projection point is inside the contour. If none of the projection points belong to the inside of the contour line on the contour map, the point is judged as a hidden point.

[0083] High-confidence points also need to be retained. High-confidence points are points with large NCC values obtained through the color correlation algorithm, that is, points with a matching metric value greater than a certain threshold. The threshold is set large enough to ensure the correctness of the points that meet the conditions. After obtaining the 3D model point cloud data, the Poisson surface reconstruction algorithm is used to triangulate the point cloud to obtain an accurate and closed mesh.

[0084] As Figure 2 shown, an embodiment of the present application further provides a 3D static model reconstruction device, including a picture acquisition and segmentation module 201, a picture stereo visualization module 202, a point cloud data acquisition module 203, and a 3D model reconstruction module 204. Among them, the picture acquisition and segmentation module 201 is used to acquire multiple picture sequences of the target object and perform binary segmentation on each picture sequence to obtain corresponding contour maps respectively; the picture stereo visualization module 202 is used to obtain a stereo visualization shell structure based on the contour maps; the point cloud data acquisition module 203 is used to acquire a number of 3D model point cloud data in each contour map in the stereo visualization shell structure based on a preset color consistency calculation criterion; the 3D model reconstruction module 204 is used to reconstruct the 3D model point cloud data.

[0085] In the picture stereo visualization module 202 of the embodiment, specifically, the visual geometric information is set based on the camera calibration principle, and the light cone region of each contour map is obtained according to the visual geometric information; the spatial intersection of all light cone regions is calculated, and the spatial intersection is defined as the stereo visualization shell structure. In the point cloud data acquisition module 203, specifically, any one contour map is selected as the current target contour map, and the color consistency between the current target contour map and the adjacent contour maps is calculated; the spatial three-dimensional point coordinates that meet the preset requirements are selected as the candidate 3D model point cloud of the current target contour map; several 3D model point clouds are selected from the candidate 3D model points as the 3D model point cloud data.

[0086] In the above, the formula for calculating the color consistency between the current target contour map and the adjacent contour maps is:

[0087]

[0088] where NCC is the color consistency; n represents the spatial intersection of n light cone regions, j represents the sequence value of pixels in a 3×3 rectangular window; p is the target pixel in the target view; s is the index number of the candidate depth; n j is the pixel in a 3×3 region block in the target view; f s (n j ) is the pixel corresponding to n j in the current target contour map, and Respectively represent the average color intensities of two regions.

[0089] The three-dimensional spatial point coordinates that meet the preset requirements are obtained through the following formula:

[0090]

[0091] Where s * is the three-dimensional spatial coordinate point.

[0092] In the three-dimensional model reconstruction module 204 of the embodiment, before reconstructing the three-dimensional model point cloud data, downsampling the three-dimensional model point cloud data is also included. Specifically, any two adjacent contour maps are selected, and the intersection points of the two adjacent contour maps are calculated; the hidden points in the three-dimensional visualization shell structure are extracted; the intersection points and hidden points are retained.

[0093] As Figure 3 shown, an embodiment of the present application also provides a computer device, including: a memory 301 and one or more processors 302; the memory 301 is used to store one or more programs; when the one or more programs are executed by the one or more processors 302, the one or more processors are enabled to implement the three-dimensional static model reconstruction method as described in the present application.

[0094] An embodiment of the present application also provides a storage medium containing computer-executable instructions, and the computer-executable instructions are used to execute the three-dimensional static model reconstruction method provided in the above embodiment when executed by a computer processor. The three-dimensional static model reconstruction method includes: collecting multiple picture sequences of a target object, and performing binary segmentation on each picture sequence to respectively obtain corresponding contour maps; obtaining a three-dimensional visualization shell structure based on the contour maps; obtaining a plurality of three-dimensional model point cloud data in each contour map of the three-dimensional visualization shell structure based on a preset color consistency calculation criterion; reconstructing the three-dimensional model point cloud data.

[0095] Storage medium - Any of various types of memory devices or storage devices. The term "storage medium" is intended to include: installation media such as CD-ROMs, floppy disks, or magnetic tape devices; computer system memory or random access memory such as DRAM, DDR RAM, SRAM, EDO RAM, Rambus RAM, etc.; non-volatile memory such as flash memory, magnetic media (such as hard disks or optical storage); registers or other similar types of memory elements, etc. The storage medium may also include other types of memory or combinations thereof. Additionally, the storage medium may be located in a first computer system in which the program is executed, or may be located in a different second computer system that is connected to the first computer system via a network (such as the Internet). The second computer system may provide program instructions to the first computer for execution. The term "storage medium" may include two or more storage media that may reside in different locations (such as in different computer systems connected via a network). The storage medium may store program instructions executable by one or more processors (such as embodied as a computer program).

[0096] Of course, for a storage medium containing computer-executable instructions provided by an embodiment of the present application, the computer-executable instructions are not limited to the path selection method based on Internet of Things data transmission as described above, and may also perform related operations in the path selection method based on Internet of Things data transmission provided by any embodiment of the present application.

[0097] The three-dimensional static model reconstruction device, equipment, and storage medium provided in the above embodiments can execute the three-dimensional static model reconstruction method provided by any embodiment of the present application. For technical details not described in detail in the above embodiments, reference may be made to the three-dimensional static model reconstruction method provided by any embodiment of the present application.

[0098] The above is only the preferred embodiment of the present application and the technical principles applied. The present application is not limited to the specific embodiments described herein. Various obvious changes, re-adjustments, and substitutions that can be made by those skilled in the art will not depart from the protection scope of the present application. Therefore, although the present application has been described in more detail through the above embodiments, the present application is not limited to the above embodiments. Without departing from the concept of the present application, it may also include more other equivalent embodiments, and the scope of the present application is determined by the scope of the claims.

Claims

1. A three-dimensional static model reconstruction method, characterized in that, Including: Collecting a plurality of picture sequences of a target object, and performing binary segmentation on each picture sequence to respectively obtain corresponding contour maps; Obtaining a three-dimensional visual hull structure based on the contour maps; Selecting any one of the contour maps as the current target contour map, and calculating the color consistency between the current target contour map and adjacent contour maps; Selecting spatial three-dimensional point coordinates that meet preset requirements as the candidate three-dimensional model point cloud of the current target contour map; Selecting several three-dimensional model point clouds from the candidate three-dimensional model point clouds as the three-dimensional model point cloud data; The formula for calculating the color consistency between the current target contour map and adjacent contour maps is: Among them, NCC is color consistency; n represents the spatial intersection of n light cone regions, j represents the sequence value of pixels in a 3×3 rectangular window; p is the target pixel; s is the index number; n j is the pixel in a 3×3 region block in the target view; f s (n j ) is the pixel corresponding to n j in the current target contour map, and represent the mean color intensities of two regions respectively; Selecting any two adjacent contour maps, and calculating the intersection points of the two adjacent contour maps; Extracting the hidden points in the three-dimensional visual hull structure; Retaining the intersection points and hidden points; Reconstructing the three-dimensional model point cloud data.

2. The three-dimensional static model reconstruction method according to claim 1, wherein The obtaining the three-dimensional visual hull structure based on the contour maps includes: Setting visual geometric information based on the camera calibration principle, and obtaining the light cone region of each contour map according to the visual geometric information; Calculating the spatial intersection of all the light cone regions, and defining the spatial intersection as the three-dimensional visual hull structure.

3. A three-dimensional static model reconstruction device, characterized in that, Including: Picture acquisition and segmentation module: for collecting a plurality of picture sequences of a target object, and performing binary segmentation on each picture sequence to respectively obtain corresponding contour maps; Picture three-dimensional visualization module: for obtaining a three-dimensional visual hull structure based on the contour maps; Point cloud data acquisition module: for selecting any one of the contour maps as the current target contour map, and calculating the color consistency between the current target contour map and adjacent contour maps; Selecting spatial three-dimensional point coordinates that meet preset requirements as the candidate three-dimensional model point cloud of the current target contour map; Selecting several three-dimensional model point clouds from the candidate three-dimensional model point clouds as the three-dimensional model point cloud data; The formula for calculating the color consistency between the current target contour map and adjacent contour maps is: Among them, NCC is color consistency; n represents the spatial intersection of n light cone regions, j represents the sequence value of pixels in a 3×3 rectangular window; p is the target pixel; s is the index number; n j is the pixel in the 3×3 region block in the target view; f s (n j ) is the pixel corresponding to n j in the current target contour map, and represent the average color intensities of the two regions respectively; Three-dimensional model reconstruction module: for selecting any two adjacent contour maps, and calculating the intersection points of the two adjacent contour maps; Extracting the hidden points in the three-dimensional visual hull structure; Retaining the intersection points and hidden points; Reconstructing the three-dimensional model point cloud data.

4. A computer device, characterized in that, Including: A memory and one or more processors; The memory is used for storing one or more programs; When the one or more programs are executed by the one or more processors, enabling the one or more processors to implement the three-dimensional static model reconstruction method as described in any one of claims 1-2.

5. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions are used for executing the three-dimensional static model reconstruction method as described in any one of claims 1-2 when executed by a computer processor.

Images