Camera calibration method, system, device, and storage medium

Abstract

The application provides a camera calibration method, system, device and storage medium, the method comprising: driving at least one target camera to focus on a target object on a turntable to obtain a focusing parameter of the at least one target camera; driving the at least one target camera to shoot a calibration body on the turntable based on the focusing parameter to obtain a first image set of image identifiers on the calibration body; and calibrating the at least one target camera according to the first image set and a preset calibration algorithm to obtain pose information of the at least one target camera relative to the target object, wherein the calibration body comprises: at least three characteristic surfaces, each of which is a concave surface; the at least three characteristic surfaces are distributed in a circumferential direction and are sequentially connected; and the at least three characteristic surfaces are respectively provided with image identifiers. The application expands the application range of the calibration body, reduces the operation process of replacing the calibration body due to different sizes of the target object, and effectively improves the accuracy of camera calibration.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image processing, and particularly relates to a camera calibration method, system, device and storage medium. BACKGROUND

[0002] Three-dimensional reconstruction refers to establishing a mathematical model suitable for computer representation and processing of a three-dimensional object, and is the basis for processing, operating and analyzing the properties of the object in a computer environment, and is also a key technology for establishing a virtual reality representing the objective world in a computer. In computer vision, three-dimensional reconstruction refers to a process of reconstructing three-dimensional information according to a single view or multiple views. Three-dimensional reconstruction has important application value in application scenarios such as AR (Augmented Reality), three-dimensional interactive display of commodities, editing and the like.

[0003] And object pose (camera pose) estimation is an indispensable input for three-dimensional reconstruction of an object. In an actual scene, because the sizes of objects to be reconstructed are often different, different sizes of calibration objects are used to calibrate a camera, and then pose information of the camera after calibration is obtained. This scheme is not only cumbersome to operate, but also easily leads to failure of camera calibration for a target object with a very small volume. SUMMARY

[0004] The main purpose of the embodiments of the present application is to provide a camera calibration method, system, device and storage medium, which realizes expansion of the application range of a calibration body, reduces the operation process of replacing the calibration body due to different sizes of target objects, and effectively improves the accuracy of camera calibration.

[0005] In a first aspect, the embodiments of the present application provide a camera calibration method, comprising: being applied to an image acquisition system, the image acquisition system comprising: a turntable for placing a target object or a calibration body, at least one target camera and the calibration body, wherein the calibration body comprises: at least three feature surfaces, the at least three feature surfaces are all concave surfaces; the at least three feature surfaces are distributed along a circumferential direction and are sequentially connected; and image identifiers are arranged on the at least three feature surfaces respectively, so that when the target camera photographs the calibration body, the image identifier on at least one feature surface of the calibration body is within the imaging range of the target camera.

[0006] The method comprises: driving the at least one target camera to focus on the target object on the turntable to obtain a focusing parameter of the at least one target camera, wherein the target object is an object that needs to be three-dimensionally reconstructed, and the target object is placed at a predetermined position on the turntable; driving the at least one target camera to capture a calibration body on the turntable based on the focusing parameter to obtain a first image set of image identifiers on the calibration body, wherein the calibration body is placed at a predetermined position on the turntable; each first image in the first image set comprises at least one image identifier of the feature surface; and calibrating the at least one target camera according to the first image set and a preset calibration algorithm to obtain pose information of the at least one target camera relative to the target object.

[0007] In an embodiment, the driving the at least one target camera to capture the calibration body on the turntable based on the focusing parameter to obtain the first image set of image identifiers on the calibration body comprises: driving the turntable to rotate at a preset rotating speed, and driving the at least one target camera to capture the calibration body on the turntable based on the focusing parameter for multiple times in the process of rotating the calibration body driven by the turntable to obtain the first image set of image identifiers on the calibration body.

[0008] In an embodiment, the driving the at least one target camera to capture the calibration body on the turntable based on the focusing parameter comprises: in the process of capturing, a central axis of the calibration body is parallel to a rotating shaft of the turntable.

[0009] In an embodiment, the feature surface comprises: at least two sub-surfaces, an included angle between two adjacent sub-surfaces is less than 180 degrees, and the included angle is an external angle of the calibration body.

[0010] In an embodiment, the calibration body comprises four feature surfaces; one feature surface comprises two sub-surfaces connected in intersection, the included angle between the two sub-surfaces is greater than or equal to 90 degrees; and the four feature surfaces are sequentially connected and distributed in a circumferential direction.

[0011] In an embodiment, the calibration body comprises eight feature surfaces; one feature surface comprises three sub-surfaces connected in intersection in pairs, the three sub-surfaces are all planes, the included angle between two adjacent sub-surfaces is 90 degrees; the three sub-surfaces form three sub-surface intersection lines, the three sub-surface intersection lines are perpendicular to each other and intersect to form a vertex of the feature surface, and the vertices of the eight feature surfaces are distributed on the same point.

[0012] In a second aspect, the embodiments of the present application provide a three-dimensional reconstruction method, applied to an image acquisition system, the image acquisition system comprising: a turntable for placing a target object or a calibration body, at least one target camera and the calibration body, wherein the calibration body comprises: at least three feature surfaces, each of the at least three feature surfaces being a concave surface; the at least three feature surfaces are distributed in a circumferential direction and are sequentially connected; image identifiers are arranged on the at least three feature surfaces respectively, so that when the target object is photographed by the target camera, the image identifier on at least one feature surface of the calibration body is within the imaging range of the target camera;

[0013] The method comprises: in response to a three-dimensional reconstruction request of a target object, driving the at least one target camera to focus on the target object on the turntable to obtain a focusing parameter of the at least one target camera, wherein the target object is an object that needs to be three-dimensionally reconstructed, and the target object is placed at a predetermined position on the turntable; driving the at least one target camera to photograph the calibration body on the turntable based on the focusing parameter to obtain a first image set of image identifiers on the calibration body, wherein the calibration body is placed at a predetermined position on the turntable; each first image in the first image set comprises at least one image identifier of the feature surface; calibrating the at least one target camera according to the first image set and a preset calibration algorithm to obtain pose information of the at least one target camera relative to the target object; driving the at least one target camera to photograph the target object on the turntable based on the focusing parameter to obtain a second image set of the target object, wherein the target object is placed at a predetermined position on the turntable; and three-dimensionally reconstructing the target object according to the second image set and the pose information to generate a three-dimensional virtual model of the target object.

[0014] In a third aspect, the embodiments of the present application provide a camera calibration device, applied to an image acquisition system, the image acquisition system comprising: a turntable for placing a target object or a calibration body, at least one target camera and the calibration body, wherein the calibration body comprises: at least three feature surfaces, each of the at least three feature surfaces being a concave surface; the at least three feature surfaces are distributed in a circumferential direction and are sequentially connected; image identifiers are arranged on the at least three feature surfaces respectively, so that when the target object is photographed by the target camera, the image identifier on at least one feature surface of the calibration body is within the imaging range of the target camera;

[0015] The device comprises:

[0016] The first driving module is configured to drive the at least one target camera to focus on the target object on the turntable to obtain a focusing parameter of the at least one target camera, wherein the target object is an object that needs to be reconstructed in three dimensions, and the target object is placed at a predetermined position on the turntable.

[0017] The second driving module is configured to drive the at least one target camera to capture the calibration body on the turntable based on the focusing parameter to obtain a first image set of image identifiers on the calibration body, wherein the calibration body is placed at a predetermined position on the turntable, and each first image in the first image set includes at least one image identifier of the feature surface.

[0018] The calibration module is configured to calibrate the at least one target camera according to the first image set and a preset calibration algorithm to obtain pose information of the at least one target camera relative to the target object.

[0019] In an embodiment, the second driving module is configured to drive the turntable to rotate at a preset rotating speed, and drive the at least one target camera to capture the calibration body on the turntable multiple times based on the focusing parameter during the process that the turntable drives the calibration body to rotate, to obtain the first image set of image identifiers on the calibration body.

[0020] In an embodiment, during the process that the at least one target camera captures the calibration body on the turntable based on the focusing parameter, a central axis of the calibration body is parallel to a rotating shaft of the turntable.

[0021] In an embodiment, the feature surface includes at least two sub-surfaces, an included angle between two adjacent sub-surfaces is less than 180 degrees, and the included angle is an external angle of the calibration body.

[0022] In an embodiment, the calibration body includes four feature surfaces, one feature surface includes two sub-surfaces that are connected in intersection, the included angle between the two sub-surfaces is greater than or equal to 90 degrees, and the four feature surfaces are sequentially connected and distributed in a circumferential direction.

[0023] In an embodiment, the calibration body includes eight feature surfaces, one feature surface includes three sub-surfaces that are connected in intersection in pairs, the three sub-surfaces are all planes, the included angle between two adjacent sub-surfaces is 90 degrees, the three sub-surfaces form three sub-surface intersection lines, the three sub-surface intersection lines are perpendicular to each other and intersect to form a vertex of the feature surface, and the vertices of the eight feature surfaces are distributed on the same point.

[0024] In a fourth aspect, an embodiment of the present application provides an image acquisition system, including:

[0025] a turntable for placing a target object or a calibration body, the target object being an object that needs to be reconstructed in three dimensions;

[0026] at least one target camera for photographing the target object or the calibration body placed on the turntable, and collecting image information of the target object or the calibration body;

[0027] The calibration body comprises:

[0028] at least three feature surfaces, each of which is a concave surface; the at least three feature surfaces are distributed in a circumferential direction and are sequentially connected; and image identifiers are arranged on the at least three feature surfaces, respectively, so that when the target camera photographs the calibration body, the image identifier on at least one of the feature surfaces of the calibration body is within the imaging range of the target camera.

[0029] In a fifth aspect, an embodiment of the present application provides an electronic device, comprising:

[0030] at least one processor; and

[0031] a memory in communication connection with the at least one processor;

[0032] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the electronic device to perform the method in any of the above aspects.

[0033] In a sixth aspect, an embodiment of the present application provides a cloud device, comprising:

[0034] at least one processor; and

[0035] a memory in communication connection with the at least one processor;

[0036] The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the cloud device to perform the method in any of the above aspects.

[0037] In a seventh aspect, an embodiment of the present application provides a computer readable storage medium, which stores computer execution instructions, and when a processor executes the computer execution instructions, the method in any of the above aspects is implemented.

[0038] In an eighth aspect, an embodiment of the present application provides a computer program product, which comprises a computer program, and when a processor executes the computer program, the method in any of the above aspects is implemented.

[0039] The camera calibration method, system, device and storage medium provided by the embodiment of the application first drive a target camera to focus on a target object placed on a turntable, fix the position and focusing parameter of the target camera, remove the target object, place a calibration body on the turntable, drive the target camera after focusing to shoot the calibration body on the turntable, and obtain a first image set of the calibration body. Since the calibration body adopts a concave structure and contains a plurality of concave feature surfaces, the concave surface of the calibration body is provided with an image mark. When the camera shoots the calibration body, compared with a convex calibration body, the surface of the concave calibration body has a larger coverage space in the depth of field direction of the camera, so that the image mark on the feature surface can cover a larger camera imaging range, and the clear image of the image mark can be obtained in the certain imaging range of the camera, the application range of the calibration body is expanded, and the operation process of replacing the calibration body due to the size difference of the target object is reduced. Moreover, the concave feature surface makes each first image contain different spatial position information, and relative spatial displacement information can be provided, the calculation difficulty of the camera calibration process is reduced, and the accuracy of the camera calibration is effectively improved. BRIEF DESCRIPTION OF DRAWINGS

[0040] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate embodiments consistent with the present application and, together with the description, further serve to explain the principles of the application. It is apparent that the accompanying drawings in the following description are some embodiments of the present application, and other drawings can be obtained according to these drawings without creative labor for those skilled in the art.

[0041] Figure 1 A structural schematic diagram of an electronic device provided by the embodiment of the application;

[0042] Figure 2 An application scene schematic diagram of a camera calibration scheme provided by the embodiment of the application;

[0043] Figure 3 An architecture schematic diagram of an image acquisition system provided by the embodiment of the application;

[0044] Figure 4 A scene schematic diagram of a camera calibration scheme provided by the embodiment of the application;

[0045] Figure 5 A flow schematic diagram of a camera calibration method provided by the embodiment of the application;

[0046] Figures 6A to 6D A structural schematic diagram of a calibration body provided by the embodiment of the application;

[0047] Figures 7A to 7B A structural schematic diagram of a calibration body provided by the embodiment of the application;

[0048] Figure 8 A flowchart of a three-dimensional reconstruction method provided by an embodiment of the present application is shown in FIG. 1.

[0049] Figure 9 A structural diagram of a camera calibration device provided by an embodiment of the present application is shown in FIG. 4.

[0050] Figure 10 A structural diagram of a cloud device provided by an embodiment of the present application is shown in FIG. 5.

[0051] The specific embodiments of the present application have been shown in the above-described drawings, and will be described in more detail hereinafter. These drawings and the written description are not intended to restrict the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0052] The exemplary embodiments will be described in detail herein with reference to the accompanying drawings. The same reference numbers in different drawings represent the same or similar elements unless otherwise represented. The embodiments described in the following exemplary embodiments do not represent all the embodiments consistent with the present application.

[0053] The term "and / or" is used herein to describe the association relationship of associated objects, and specifically represents that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone.

[0054] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions, and provide corresponding operation portal for user to choose authorization or refusal.

[0055] In order to clearly describe the technical solutions of the embodiments of the present application, first, the terms involved in the present application are explained:

[0056] 3D: 3-dimension, three-dimensional.

[0057] AR: Augmented Reality, augmented reality.

[0058] SFM: Structure From Motion, is a technique for estimating three-dimensional structure from a sequence of multiple two-dimensional images containing visual motion information.

[0059] For example, the user information and data involved in the present application include but are not limited to the following information and data: Figure 1As shown, the embodiment provides an electronic device 10, comprising at least one processor 110 and a memory 120, Figure 1 The processor 110 and the memory 120 are connected through the bus 100. The memory 120 stores instructions executable by the processor 110, and the instructions are executed by the processor 110 to enable the electronic device 10 to execute all or part of the processes of the method in the following embodiments, so as to effectively improve the accuracy of camera calibration and further improve the precision of three-dimensional reconstruction.

[0060] In an embodiment, the electronic device 10 can be a mobile phone, a tablet computer, a notebook computer, a desktop computer, or a large-scale computing system composed of multiple computers.

[0061] Figure 2 A schematic diagram of an application scenario system 200 of three-dimensional reconstruction provided by the embodiment is shown in the figure. Figure 2 As shown, the system comprises a server 210 and a terminal 220, wherein:

[0062] The server 210 can be a cloud data platform providing three-dimensional reconstruction services, such as an e-commerce shopping platform. In actual scenarios, an e-commerce shopping platform can have multiple servers 210, Figure 2 In an embodiment, one server 210 is taken as an example.

[0063] The terminal 220 can be a computer, a mobile phone, a tablet computer, etc. used by a user when logging into an e-commerce shopping platform, and the terminal 220 can also have multiple, Figure 2 In an embodiment, one terminal 220 is taken as an example.

[0064] The image collector 230 can be a separate camera or a camera embedded in a terminal device, and the image collector can have multiple, Figure 2 In an embodiment, one image collector 230 is taken as an example.

[0065] The terminal 220, the image collector 230, and the server 210 can transmit information through the Internet or other communication methods, so that the terminal 220 and the image collector 230 can access each other's data, and the terminal 220 and the image collector 230 can access the data on the server 210. The terminal 220 and / or the server 210 described above can be implemented by the electronic device 10.

[0066] The method provided by the embodiment can be implemented by the electronic device 10 executing corresponding software codes, and the method is realized by data interaction with the server. The electronic device 10 can be a local terminal device, such as Figure 2The method can be implemented and executed based on a cloud interaction system when the method runs on the server, where the cloud interaction system includes the server and the client device.

[0067] In a possible implementation, the method provided by the embodiment of the application provides a graphical user interface by a terminal device, where the terminal device can be the local terminal device mentioned above or the client device in the cloud interaction system mentioned above.

[0068] With the development of e-commerce shopping, more and more users choose to select goods on an e-commerce shopping platform. Some goods need to be matched with the user's own conditions for use, such as shoes and clothes. The buyer user hopes to be able to select shoes or clothes that match their own size. With the development of AR technology, using AR technology to display product information and realize interaction with users has become a trend. For example, the on-sale shoes can be three-dimensionally reconstructed to generate a 3D model, which is displayed on the e-commerce shopping platform for the buyer user to try on online based on the AR technology.

[0069] Three-dimensional reconstruction refers to establishing a mathematical model suitable for computer representation and processing of a three-dimensional object, which is the basis for processing, operating and analyzing the properties of the object in a computer environment, and is also a key technology for establishing a virtual reality in a computer to express the objective world. In computer vision, three-dimensional reconstruction refers to the process of reconstructing three-dimensional information according to a single view or multiple views of images. Three-dimensional reconstruction has important application value in AR (Augmented Reality), three-dimensional interactive display of goods, editing and other application scenarios.

[0070] And object pose (camera pose) estimation is an indispensable input for three-dimensional reconstruction of an object. In actual scenarios, because the sizes of objects to be reconstructed are often different, different sizes of calibration objects are used to calibrate the camera to obtain the pose information of the calibrated camera. This scheme is not only cumbersome to operate, but also easily leads to failure of camera calibration for very small target objects.

[0071] As shown in FIG. 1, an image acquisition system 300 provided by an embodiment of the application includes: Figure 3 A turntable 310 is configured to place a target object or a calibration body 320.

[0072] Figure 3 For example, the calibration body 320 is placed on the turntable 310. The target object is an object to be three-dimensionally reconstructed.

[0073] ​At least one target camera 330 is configured to capture images of the target object or the calibration body 320 placed on the turntable 310. In actual scenarios, multiple target cameras 330 can be arranged at different positions to capture images of the target object or the calibration body 320 from different angles. For example, an arc-shaped support can be arranged in the vertical direction along the rotation axis of the turntable 310, and multiple target cameras 330 can be arranged at different positions on the arc-shaped support. In this way, each target camera 330 can capture images of the horizontally rotating calibration body 320 or target object at different vertical positions during the rotation of the turntable 310, thereby enriching the spatial information of the captured images. Figure 3 The target camera 330 is taken as an example.

[0074] During the three-dimensional reconstruction of the target object, the camera of the image acquisition system 300 needs to be calibrated first to obtain the pose information of the camera, and then the images of the target object are acquired based on the pose information for three-dimensional reconstruction. During the calibration of the camera, the target object to be reconstructed is placed on the turntable 310, and the target camera 330 is focused on the target object on the turntable 310 to obtain the focusing parameters of the target camera 330. Then, the focusing parameters of the target camera 330 are fixed, the focal length of the focused target camera 330 is determined, and the imaging range is determined. After that, the target object is removed from the turntable 310, and the calibration body 320 is placed on the turntable 310. The focused target camera 330 captures the calibration body 320 in the state to obtain the image of the calibration body 320, and the camera calibration is performed based on the image of the calibration body 320.

[0075] In actual scenarios, the calibration body 320 can generally be formed by attaching multiple calibration boards to the surface of a cubic object. Therefore, the size of one calibration body 320 is fixed, and the surface of the calibration body 320 is generally a plane or an outwardly convex surface. Figure 4As shown, it is a contrast diagram of large-size target object A1 and small-size target object A1. Assuming that the size of the target object A1 to be reconstructed is large, the target camera 330 is first focused based on the large-size target object A, and the imaging range is fixed. Therefore, the size of the calibration body B1 should be similar to that of the target object A1, so that a relatively clear image of the calibration body B1 can be collected. At this time, if the size of the next target object A2 to be reconstructed is relatively small compared with the target object A1, the target camera 330 will first focus based on the small-size target object A2, and the imaging range will be fixed according to the small-size target object A2. If the large-size calibration body B1 is still used to calibrate the camera, the calibration body B1 will be too close to the target camera 330 at this time, and will not be in the imaging range of the target camera 330, such as exceeding the depth of field range of the target camera 330. Therefore, the image of the calibration body B1 taken by the target camera 330 at this time will be blurred, which will lead to inaccurate camera calibration. Therefore, in the related art, the calibration body B2 can be replaced, and the size of the calibration body B2 is similar to that of the target object A2. In this way, the clarity of the collected image can be improved as much as possible. In this way, different sizes of calibration bodies need to be replaced when shooting different types of target objects, which is not only troublesome to operate, but also easy to cause calibration failure when shooting small objects.

[0076] In addition, the calibration body 320 of the convex polyhedron such as a cube will have the phenomenon that the image taken only contains a single plane. This phenomenon will lead to difficulty in calculation of the calibration algorithm, such as the calculation of the fundamental matrix in the SFM algorithm, which requires relative spatial displacement information in the image. However, a single plane cannot provide an image containing relative spatial displacement, thereby increasing the difficulty of calculating the fundamental matrix.

[0077] To solve the above problems, an embodiment of the present application provides a camera calibration scheme, wherein the calibration body 320 comprises:

[0078] At least three feature surfaces, each of which is a concave surface. The at least three feature surfaces are distributed in a circumferential direction and are connected in sequence. Image identifiers are arranged on the at least three feature surfaces, respectively, so that when the target camera 330 shoots the calibration body 320, the image identifier on at least one of the feature surfaces of the calibration body 320 is in the imaging range of the target camera 330.

[0079] The calibration body 320 of the embodiment of the present application adopts a concave structure, includes a plurality of concave feature surfaces, and the concave surface of the calibration body 320 is provided with an image mark. When the camera shoots the calibration body 320, compared with the convex calibration body 320, the surface of the concave calibration body 320 has a larger coverage space in the depth of field direction of the camera, so that the image mark on the feature surface can cover a larger camera imaging range. The clear image of the image mark can be obtained in the certain imaging range of the camera, the application range of the calibration body 320 is expanded, the operation process of replacing the calibration body 320 caused by different sizes of target objects is reduced. Moreover, the concave feature surface makes each first image contain different spatial position information, thereby providing relative spatial displacement information, reducing the calculation difficulty of the camera calibration process, and effectively improving the accuracy of the camera calibration.

[0080] Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and the features in the embodiments can be combined with each other without conflict. In addition, the sequence of steps in each method embodiment is only an example, not a strict limitation.

[0081] Please refer to Figure 5 , which is a camera calibration method of an embodiment of the present application. The method can be executed by the electronic device 10 shown in Figure 1 , and can be applied to the camera calibration application scenario shown in Figures 2 to 4 , to expand the application range of the calibration body 320, reduce the operation process of replacing the calibration body 320 caused by different sizes of target objects, and effectively improve the accuracy of the camera calibration. The terminal 220 is taken as an example of the execution end in the embodiment, and the method includes the following steps:

[0082] Step 501: Drive at least one target camera 330 to focus on a target object on the turntable 310, and obtain the focusing parameter of the at least one target camera 330. The target object is placed at a predetermined position on the turntable 310.

[0083] In this step, the target object is an object that needs to be reconstructed in three dimensions, such as a packaging box, a shoe, a bottle, a cup, etc. Taking an e-commerce scenario as an example, the target object can be a commodity that needs to realize AR interaction, such as a shoe, a hat, etc. Before three-dimensional reconstruction of the target object, the target camera 330 needs to be calibrated first to obtain the intrinsic and extrinsic parameters of the camera. In the camera calibration process, the target object can be placed at a predetermined position on the turntable 310 first, and then the target camera 330 is driven to focus on the target camera 330 on the turntable 310. At this time, the turntable 310 can remain stationary, and the focusing parameter of the target camera 330 at the current relative position can be obtained. The focusing parameter can include the focal point and the depth of field of the target camera 330, etc. The predetermined position can be set based on actual needs, and the predetermined position is used to ensure that the target object and the calibration body 320 can be clearly imaged under the same focusing parameter after focusing, for example, the predetermined position can be the center point position of the turntable 310.

[0084] Step 502: Drive at least one target camera 330 to capture the calibration body 320 on the turntable 310 based on the focusing parameter to obtain a first image set of the image identifier on the calibration body 320, wherein the calibration body 320 is placed at a predetermined position on the turntable 310. Each first image in the first image set includes at least one image identifier of a feature surface.

[0085] In this step, when the target camera 330 is focused, the target object is removed from the turntable 310, and the calibration body 320 is placed at a predetermined position on the turntable 310. At this time, the target camera 330 can be driven to capture the calibration body 320 according to the focusing parameter after focusing in step 501 to obtain a plurality of first images of the calibration body 320, forming a first image set. The first image includes at least an image identifier of a feature surface on the calibration body 320. Since the feature surface of the calibration body 320 is a concave surface, different spatial positions can be captured in the first image. The image identifier is used as input data for the camera calibration algorithm, which can provide rich relative spatial position information and reduce the difficulty of the algorithm. The image identifier can be a preset pattern identifier or a coding-supported pattern identifier. The style of the image identifier is not limited in this embodiment. The image identifier can be distributed on all visible surfaces of the calibration body 320. The size of the calibration body 320 can be matched with the size of the turntable 310.

[0086] Please refer to Figure 6A which is a structural schematic diagram of the calibration body 320 of an embodiment of the present application. The calibration body 320 can be used in Figure 3 the image acquisition system 300 shown in the figure for calibrating the target camera 330. The calibration body 320 can specifically include:

[0087] There are at least three feature surfaces, and all three feature surfaces are concave. The at least three feature surfaces are distributed circumferentially and connected sequentially. For example... Figure 6A As shown, taking three feature surfaces as an example, namely feature surface 1, feature surface 2, and feature surface 3, feature surface 1 and feature surface 2 intersect and connect to form intersection line S1, feature surface 2 and feature surface 3 intersect and connect to form intersection line S2, and feature surface 3 and feature surface 1 intersect and connect to form intersection line S3. In this case, the calibration body 320 can be a three-concave prism, and the center of the prism can be hollow or solid. The material of the calibration body 320 can be solid wood, plastic, metal, or other materials that can maintain a stable structure. This embodiment does not limit the material of the calibration body 320.

[0088] Image markers are provided on at least three feature surfaces, so that when the target camera 330 takes a picture of the calibration body 320, at least one image marker on a feature surface of the calibration body 320 is within the imaging range of the target camera 330.

[0089] Since the calibration body 320 is composed of multiple concave feature surfaces, and image markers are provided on the concave surfaces of the calibration body 320, when the camera captures an image of the calibration body 320, compared to flat or convex surfaces, the concave calibration body 320 has a larger coverage space in the camera's depth of field direction. Therefore, the image markers on the feature surface can cover a larger camera imaging range. This allows the camera to obtain clear images of the image markers within a larger imaging range, expanding the applicability of the calibration body 320 and reducing the need to replace the calibration body 320 due to different target object sizes. Furthermore, each first image contains image markers at different spatial positions, thus providing relative spatial displacement information, reducing the computational difficulty of the camera calibration process, and effectively improving the efficiency of camera calibration. In one embodiment, the feature surface includes at least two sub-surfaces, with the included angle between two adjacent sub-surfaces being less than 180 degrees, and the included angle being an exterior angle of the calibration body 320.

[0090] In this embodiment, as Figure 6BAs shown, taking the example that the feature surface 1 includes two sub-surfaces, the feature surface 1 can include a sub-surface 11 and a sub-surface 12, the sub-surface 11 and the sub-surface 12 are connected in intersection, forming an intersection line S101, the angle between the sub-surface 11 and the sub-surface 12 is an outward angle, the angle faces the target camera 330 when the target camera 330 shoots the calibration body 320, and the angle is less than 180 degrees, so as to ensure that the feature surface 1 formed after the sub-surface 11 and the sub-surface 12 are connected in intersection is a concave surface. Image identifiers can be arranged on each sub-surface, so that the target camera 330 can shoot the image identifiers on different sub-surfaces from multiple perspectives, so that each image in the first image set includes image identifiers on different sub-surfaces, and then the relative spatial position information between different sub-surfaces can be represented, accurate input data is provided for the calibration algorithm, the parameters of the target camera 330 are calculated by the preset calibration algorithm, and then the accuracy of the camera calibration result is improved.

[0091] In an embodiment, the calibration body 320 includes four feature surfaces. One of the feature surfaces includes two sub-surfaces connected in intersection, and the angle between the two sub-surfaces is greater than or equal to 90 degrees. The four feature surfaces are sequentially connected and distributed in a circumferential direction.

[0092] In the embodiment, as shown, Figure 6C taking the example of the feature surface 1, the feature surface 1 includes a sub-surface 11 and a sub-surface 12, and image identifiers are arranged on each sub-surface. The four feature surfaces formed are all concave surfaces, so that the calibration body 320 is a concave polyhedron. When the camera shoots the calibration body 320, it can be ensured that the image identifiers on different sub-surfaces can be shot from multiple perspectives, so that each image in the first image set includes image identifiers on different sub-surfaces, and then the relative spatial position information between different sub-surfaces can be represented, accurate input data is provided for the calibration algorithm, the parameters of the target camera 330 are calculated by the preset calibration algorithm, and then the accuracy of the camera calibration result is improved.

[0093] In an actual scene, as shown, Figure 6D the angle between the sub-surface 11 and the sub-surface 12 can be 90 degrees, so that the concave surface of the calibration body 320 has a larger coverage range along the depth of field direction of the target camera 330, and then the calibration body 320 is suitable for pose calibration of a target object with a larger size variation range, the use range of the calibration body 320 is expanded, and the operation of replacing the calibration body 320 due to different sizes of the target object is reduced.

[0094] In an embodiment, the calibration body 320 includes eight feature surfaces. One of the feature surfaces includes three sub-surfaces connected in intersection, the three sub-surfaces are all planes, and the angle between adjacent two sub-surfaces can be 90 degrees. The three sub-surfaces form three sub-surface intersection lines, the three sub-surface intersection lines are perpendicular to each other and intersect to form the vertex of the feature surface, and the vertices of the eight feature surfaces are distributed on the same point.

[0095] In the embodiment, as shown in Figure 7A The characteristic surface 1 includes three planes, i.e., the sub-surface 11, the sub-surface 12 and the sub-surface 13, and each of the sub-surfaces is provided with an image identifier. The eight characteristic surfaces are all concave, so that the calibration body 320 is a concave polyhedron, and the eight characteristic surfaces can cover a larger visual angle range of the target camera 330, so that when the camera captures the calibration body 320, more visual angles can be ensured to capture the image identifiers on different sub-surfaces, so that each image in the first image set contains image identifiers on different sub-surfaces, and thus the relative spatial position information between different sub-surfaces can be represented, accurate input data for the calibration algorithm is provided, the parameters of the target camera 330 are effectively assisted to be calculated by the preset calibration algorithm, and thus the accuracy of the camera calibration result is improved.

[0096] In an actual scene, as shown in Figure 7B The angles between adjacent two of the sub-surface 11, the sub-surface 12 and the sub-surface 13 can be 90 degrees, i.e., the sub-surface 11, the sub-surface 12 and the sub-surface 13 are perpendicular to each other, the intersection points form the right-angle polyhedron vertices of the characteristic surface 1, and the eight characteristic surfaces form eight vertices, which coincide, so that the concave surface of the calibration body 320 has a larger coverage range along the depth of field direction of the target camera 330, and thus the calibration body 320 is suitable for pose calibration of a target object with a larger size variation range, the use range of the calibration body 320 is expanded, and the operation of replacing the calibration body 320 due to different sizes of the target object is reduced. For example, after the target camera 330 focuses on a very small target object (such as a ring), the focal point is about the center point of the ring, and after the calibration body 320 is placed at a predetermined position on the turntable 310, the vertices of the eight characteristic surfaces are also located near the focal point. When the target camera 330 captures the calibration body 320 in the embodiment, a clear image of the calibration body 320 can still be captured, the image identifiers are arranged on each sub-surface of the calibration body 320, and thus the image identifiers of different sub-surfaces can be captured, without the need to replace the calibration body 320 with the same size as the ring, and the calibration failure phenomenon caused by the small size of the calibration body 320 is avoided.

[0097] Compared with the convex polyhedron calibration body 320 such as a cube, the calibration body 320 in the embodiment does not have the phenomenon that only a single plane is contained in the captured image. Each first image in the first image set captured by the calibration body 320 in the application contains different spatial position information, so that the relative spatial displacement information can be provided, the calculation difficulty of the fundamental matrix is reduced, and the efficiency of camera calibration is improved.

[0098] In an embodiment, step 502 can specifically include: driving the turntable 310 to rotate at a preset rotating speed, and driving the at least one target camera 330 to capture the calibration body 320 on the turntable 310 multiple times based on the focusing parameter during the rotation of the calibration body 320 driven by the turntable 310, to obtain a first image set of the image mark on the calibration body 320.

[0099] In the embodiment, during the collection of the first image set of the calibration body 320, the turntable 310 can be set to rotate at a preset rotating speed, and the target camera 330 can be fixed. During the rotation of the turntable 310, the target camera 330 can be driven to capture the calibration body 320 on the turntable 310 multiple times according to the focusing parameter every preset time period, to obtain the first image set of the image mark on the calibration body 320, so as to ensure that the first image set contains images of the calibration body 320 at different angles of view, and to obtain more abundant calibration input data. For example, the shooting interval of the turntable 310 and the target camera 330 can be matched with each other, so that the target camera 330 can capture 72 first images when the calibration body 320 rotates one round with the turntable 310, i.e., one image is captured every 5 degrees for one round of the calibration body 320, to obtain the first image set.

[0100] In an embodiment, step 502 can specifically include: during the shooting, the central axis of the calibration body 320 is parallel to the rotating shaft of the turntable 310.

[0101] In the embodiment, in order to ensure that the feature surfaces of the calibration body 320 can all be within the imaging range of the target camera 330, the central axis of the calibration body 320 can be placed parallel to the rotating shaft of the turntable 310, for example, the central axis of the calibration body 320 can be placed coincident with the rotating shaft of the turntable 310. Since the feature surfaces of the calibration body 320 are distributed along the circumferential direction, at least one feature surface can face the target camera 330 during the shooting, to ensure that the effective first images are collected.

[0102] Step 503: calibrating the at least one target camera 330 according to the first image set and a preset calibration algorithm, to obtain camera pose information of the at least one target camera 330 relative to the target object.

[0103] In this step, the preset calibration algorithm can be an SFM algorithm or other calibration algorithms, and the embodiment does not limit the type of preset calibration algorithm. The first image set is taken as the input of the preset calibration algorithm, the image features of the first image are extracted based on the preset calibration algorithm, and then image matching is performed based on the image features. Based on the image matching result, the camera pose information of the target camera 330 relative to the target object is determined, which can include camera intrinsic and extrinsic parameters, such as camera focal point, image principal point coordinates, rotation matrix, and translation matrix parameters. Since each first image in the first image set contains different spatial position information, it can provide relative spatial displacement information, reduce the calculation difficulty of the camera calibration process, and thus effectively improve the efficiency of camera calibration.

[0104] The above camera calibration method first drives the target camera 330 to focus on the target object placed on the turntable 310, fixes the position and focusing parameters of the target camera 330, removes the target object, places the calibration body 320 on the turntable 310, and drives the focused target camera 330 to shoot the calibration body 320 on the turntable 310 to obtain a first image set of the calibration body 320. Since the calibration body 320 adopts an inner concave structure and contains multiple concave feature surfaces, the concave surface of the calibration body 320 is provided with image identifiers. When the camera shoots the calibration body 320, compared with the convex calibration body 320, the surface of the concave calibration body 320 has a larger coverage space in the camera depth of field direction, so the image identifiers on the feature surface can cover a larger camera imaging range. The camera can obtain clear images of the image identifiers within a certain imaging range, expand the application range of the calibration body 320, and reduce the operation process of replacing the calibration body 320 due to different sizes of target objects.

[0105] Please refer to Figure 8 which is a three-dimensional reconstruction method of an embodiment of the present application. The method can be executed by the electronic device 10 shown in Figure 1 , and can be applied to the application scenarios shown in Figures 2 to 4 to reduce the operation process of replacing the calibration body 320 due to different sizes of target objects, effectively improve the accuracy of camera calibration, and further improve the efficiency of three-dimensional reconstruction. The embodiment takes the terminal 220 as an execution end as an example, and compared with the foregoing embodiment, the embodiment takes a three-dimensional reconstruction scene of a target object as an example. The method includes the following steps:

[0106] Step 801: in response to a three-dimensional reconstruction request for a target object, driving at least one target camera 330 to focus on the target object on the turntable 310, to obtain the focusing parameter of the at least one target camera 330, wherein the target object is an object that needs to be three-dimensionally reconstructed, and the target object is placed at a predetermined position on the turntable 310.

[0107] In this step, the user can trigger the three-dimensional reconstruction request for the target object, or the three-dimensional reconstruction request for the target object can be triggered automatically by a predetermined condition. For example, in an e-commerce scenario, a merchant triggers a three-dimensional reconstruction request for a certain shoe. In response to the three-dimensional reconstruction request, the image acquisition system 300 described above can be used to acquire images of the target object. Before acquiring the images, the target camera 330 of the image acquisition system 300 is calibrated first to improve the accuracy of the image acquisition result, thereby providing accurate input data for three-dimensional reconstruction. For details, see the related description of step 501 in the foregoing embodiments.

[0108] Step 802: driving at least one target camera 330 to capture the calibration body 320 on the turntable 310 based on the focusing parameter, to obtain a first image set of the image identifier on the calibration body 320, wherein the calibration body 320 is placed at a predetermined position on the turntable 310. Each first image in the first image set includes at least one image identifier of a feature surface. For details, see the related description of step 502 in the foregoing embodiments.

[0109] Step 803: calibrating at least one target camera 330 according to the first image set and a preset calibration algorithm, to obtain the pose information of at least one target camera 330 relative to the target object. For details, see the related description of step 503 in the foregoing embodiments.

[0110] Step 804: driving at least one target camera 330 to capture the target object on the turntable 310 based on the focusing parameter, to obtain a second image set of the target object, wherein the target object is placed at a predetermined position on the turntable 310.

[0111] In this step, after step 803, the camera of the image acquisition system 300 completes the calibration process relative to the target object, the calibration body 320 can be removed from the turntable 310, and the target object that needs to be three-dimensionally reconstructed is placed at a predetermined position on the turntable 310, such as a packaging bottle that needs to be three-dimensionally reconstructed. The turntable 310 is driven to rotate at a preset speed, and the target camera 330 captures the packaging bottle multiple times during the rotation of the packaging bottle with the turntable 310, such as capturing a second image every 5 degrees. A total of 72 second images of the packaging bottle are captured in one rotation, to obtain a second image set. In actual scenarios, the number of second images can be set according to actual needs.

[0112] Step 805: three-dimensional reconstruction of the target object is performed according to the second image set and the pose information, and a three-dimensional virtual model of the target object is generated.

[0113] In this step, the multi-view images of the target object included in the second image set can be stereo matched. For example, based on the SFM algorithm, matching feature points can be calculated for each two adjacent images. Generally, two images are first used for reconstruction, and an initial point cloud is calculated. Then, subsequent images are continuously added. The method of adding which image can be: check which image in the existing images matches the most with the points in the existing point cloud, and select which image for reconstruction. After calculating the matching feature points, the target object can be three-dimensionally reconstructed in combination with the pose information obtained in step 803, and a three-dimensional virtual model of the target object is obtained.

[0114] For example, the target object is a packaging bottle in an e-commerce scenario. The packaging bottle can be three-dimensionally reconstructed based on the above data to generate a three-dimensional virtual model of the packaging bottle, and the three-dimensional virtual model of the packaging bottle can be put into an e-commerce shopping platform for buyers to view and try on, thereby improving the user's interactive experience.

[0115] The embodiments of the present application can be applied to scenarios including but not limited to AR augmented reality, object 3D interactive display, object 3D model editing, etc., and can improve the success rate of object three-dimensional reconstruction and improve the user's interactive experience.

[0116] The steps of the above three-dimensional reconstruction method can be referred to the related descriptions of the foregoing embodiments, which will not be repeated here.

[0117] Please refer to Figure 9 which is a camera calibration device 900 of an embodiment of the present application. The device can be applied to Figure 1 the electronic device 10 shown. And can be applied to the foregoing image acquisition system to achieve the following effects: expanding the application range of the calibration body, reducing the operation process of replacing the calibration body due to different sizes of target objects, and effectively improving the accuracy of camera calibration. The device includes a first driving module 901, a second driving module 902, and a calibration module 903. The functions and principles of each module are as follows:

[0118] The first driving module 901 is configured to drive at least one target camera to focus on a target object on a turntable to obtain a focusing parameter of the at least one target camera, wherein the target object is an object that needs to be three-dimensionally reconstructed, and the target object is placed at a predetermined position on the turntable.

[0119] The second driving module 902 is used to drive at least one target camera to capture images of a calibration body on a turntable based on focusing parameters, thereby obtaining a first set of images of image markers on the calibration body, wherein the calibration body is placed at a predetermined position on the turntable. Each first image in the first set includes at least one image marker of a feature surface.

[0120] The calibration module 903 is used to calibrate at least one target camera according to the first image set and the preset calibration algorithm to obtain the pose information of at least one target camera relative to the target object.

[0121] In one embodiment, the second driving module 902 is used to drive the turntable to rotate at a preset speed. During the process of the turntable driving the calibration body to rotate, at least one target camera is driven to take multiple pictures of the calibration body on the turntable based on the focusing parameters to obtain a first image set of image markers on the calibration body.

[0122] In one embodiment, during the process of driving at least one target camera to capture images of a calibration body on a turntable based on focus parameters, the central axis of the calibration body is parallel to the rotation axis of the turntable.

[0123] In one embodiment, the feature surface includes at least two sub-faces, wherein the included angle between two adjacent sub-faces is less than 180 degrees, and the included angle is an exterior angle of the calibration body.

[0124] In one embodiment, the calibration body includes four feature surfaces. Each feature surface includes two intersecting sub-surfaces, the included angle between the two sub-surfaces being greater than or equal to 90 degrees. The four feature surfaces are connected sequentially and distributed circumferentially.

[0125] In one embodiment, the calibration body includes eight feature surfaces. Each feature surface includes three intersecting sub-faces, all of which are planar, with an included angle of 90 degrees between adjacent sub-faces. The three sub-faces form three intersecting lines, which are perpendicular to each other and intersect to form the vertices of the feature surface. The vertices of the eight feature surfaces are distributed at the same point.

[0126] For a detailed description of the camera calibration device 900 described above, please refer to the description of the relevant method steps in the above embodiments. The implementation principle and technical effect are similar, and will not be repeated here in this embodiment.

[0127] Figure 10 This is a schematic diagram of the structure of a cloud device 100 provided as an exemplary embodiment of this application. The cloud device 100 can be used to run the methods provided in any of the above embodiments. Figure 10 As shown, the cloud device 100 may include: a memory 1004 and at least one processor 1005. Figure 10 Let's take a processor as an example.

[0128] The memory 1004 is configured to store computer programs and can be configured to store other various data to support operations on the cloud device 100. The memory 1004 can be an Object Storage Service (OSS).

[0129] The memory 1004 can be implemented by any type of volatile or nonvolatile storage devices or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0130] The processor 1005 is coupled to the memory 1004 and configured to execute computer programs in the memory 1004, so as to implement the solutions provided by any of the above method embodiments. The specific functions and technical effects that can be achieved are not described here.

[0131] Further, as shown in Figure 10 The cloud device further includes a firewall 1001, a load balancer 1002, a communication component 1006, a power supply component 1003 and other components. Figure 10 Some components are only schematically shown in the cloud device, which does not mean that the cloud device only includes Figure 10 the components shown.

[0132] In an embodiment, the communication component 1006 in the above Figure 10 The communication component 1006 is configured to facilitate wired or wireless communication between the device where the communication component 1006 is located and other devices. The device where the communication component 1006 is located can access a wireless network based on a communication standard, such as WiFi, 2G, 3G, 4G, LTE (Long Term Evolution, LTE for short), 5G, or a combination thereof. In an example embodiment, the communication component 1006 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an example embodiment, the communication component 1006 further includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on Radio Frequency Identification (RFID) technology, Infrared Data Association (IrDA) technology, Ultra Wide Band (UWB) technology, Bluetooth (BT) technology and other technologies.

[0133] In one embodiment, the above Figure 10 The power supply component 1003 provides power to various components of the device in which it resides. The power supply component 1003 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the device in which the power supply component resides.

[0134] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the method of any of the foregoing embodiments.

[0135] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the method of any of the foregoing embodiments.

[0136] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and there may be other division methods in actual implementation. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed.

[0137] The integrated modules described above, implemented as software functional modules, can be stored in a computer-readable storage medium. These software functional modules, stored in a storage medium, include several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) or processor to execute some steps of the methods of the various embodiments of this application.

[0138] It should be understood that the aforementioned processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. A general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in the application can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor. The memory may include high-speed RAM (Random Access Memory), and may also include non-volatile memory (NVM), such as at least one disk storage device, and may also be a USB flash drive, external hard drive, read-only memory, disk, or optical disc, etc.

[0139] The storage medium can be implemented by any type of volatile or non-volatile storage devices or a combination thereof, such as static random-access memory (SRAM), electrically erasable programmable read only memory (EEPROM), erasable programmable read only memory (EPROM), programmable read only memory (PROM), read only memory (ROM), magnetic storage, flash memory, magnetic or optical disk. The storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0140] An exemplary storage medium is coupled to the processor such that the processor can read information from, and write information to, the storage medium. Of course, the storage medium can be part of the processor. The processor and the storage medium can be located in an application specific integrated circuit (ASIC). The processor and the storage medium can be located in a remote terminal, a server, a client, or any other device in a distributed system.

[0141] It should be noted that, in the present document, the terms "comprising", "containing" or any other variant thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or apparatus that comprises a list of elements does not only include those elements, but can also include other elements not expressly listed or inherent to such process, method, article or apparatus. Without more limitations, an element defined by the phrase "comprising a" does not exclude the presence of additional identical elements in the process, method, article or apparatus that includes the element.

[0142] The above-mentioned sequence number of the embodiments of the present application is only for description, and does not represent the advantages and disadvantages of the embodiments.

[0143] Through the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned example method can be realized by means of software and a necessary general hardware platform, of course, it can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product in essence or in the form of a contribution to the prior art. The computer software product is stored in a storage medium (such as a ROM / RAM, a magnetic disk, an optical disk), and includes a plurality of instructions for causing a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the method of each embodiment of the present application.

[0144] In the technical solutions of the present application, the collection, storage, use, processing, transmission, provision and disclosure of user data and other information comply with relevant laws and regulations and do not violate public order and good customs.

[0145] The above is only a preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation using the content of the present application specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A camera calibration method, characterized in that, An image acquisition system is used, comprising: a turntable for placing a target object or calibration body; at least one target camera and a calibration body; wherein the calibration body comprises: at least three feature surfaces, all of which are concave; the at least three feature surfaces are distributed circumferentially and connected sequentially; and image markers are respectively provided on the at least three feature surfaces, so that when the target camera captures an image of the calibration body, at least one image marker on the calibration body is within the imaging range of the target camera; The method includes: Drive the at least one target camera to focus on the target object on the turntable to obtain the focus parameters of the at least one target camera, wherein the target object is an object that needs to be reconstructed in three dimensions, and the target object is placed at a predetermined position on the turntable; The at least one target camera is driven to capture images of a calibration body on the turntable based on the focusing parameters, thereby obtaining a first set of images of image markers on the calibration body, wherein the calibration body is placed at a predetermined position on the turntable; each first image in the first set of images includes at least one image marker of the feature surface; The at least one target camera is calibrated based on the first image set and a preset calibration algorithm to obtain the pose information of the at least one target camera relative to the target object.

2. The method according to claim 1, characterized in that, The process of driving the at least one target camera to capture images of the calibration object on the turntable based on the focusing parameters, thereby obtaining a first set of images of image markers on the calibration object, including: The turntable is driven to rotate at a preset speed. During the rotation of the calibration body by the turntable, at least one target camera is driven to take multiple pictures of the calibration body on the turntable based on the focusing parameters, thereby obtaining a first set of images of the image markers on the calibration body.

3. The method according to claim 1 or 2, characterized in that, The step of driving the at least one target camera to capture images of the calibration object on the turntable based on the focusing parameters includes: During the shooting process, the central axis of the calibration body is parallel to the rotation axis of the turntable.

4. The method according to claim 1, characterized in that, The feature surface includes at least two sub-faces, the included angle between two adjacent sub-faces being less than 180 degrees, and the included angle being the exterior angle of the calibration body.

5. The method according to claim 4, characterized in that, The calibration body includes the four aforementioned feature surfaces; One of the feature surfaces includes: two intersecting and connected sub-surfaces, wherein the included angle between the two sub-surfaces is greater than or equal to 90 degrees; The four feature surfaces are connected in sequence and distributed circumferentially.

6. The method according to claim 4, characterized in that, The calibration body includes eight of the aforementioned feature surfaces; One of the feature surfaces includes: three sub-surfaces that are connected in pairs, all of which are planes, and the included angle between two adjacent sub-surfaces is 90 degrees; the three sub-surfaces form three sub-surface intersection lines, which are perpendicular to each other and intersect to form the vertex of the feature surface, and the vertices of the eight feature surfaces are distributed at the same point.

7. A three-dimensional reconstruction method, characterized in that, An image acquisition system is used, comprising: a turntable for placing a target object or calibration body; at least one target camera and a calibration body; wherein the calibration body comprises: at least three feature surfaces, all of which are concave; the at least three feature surfaces are distributed circumferentially and connected sequentially; and image markers are respectively provided on the at least three feature surfaces, so that when the target camera captures an image of the calibration body, at least one image marker on the calibration body is within the imaging range of the target camera; The method includes: In response to a request for 3D reconstruction of a target object, the at least one target camera is driven to focus on the target object on the turntable to obtain the focus parameters of the at least one target camera, wherein the target object is an object that needs to be 3D reconstructed and the target object is placed at a predetermined position on the turntable. The at least one target camera is driven to capture images of a calibration body on the turntable based on the focusing parameters, thereby obtaining a first set of images of image markers on the calibration body, wherein the calibration body is placed at a predetermined position on the turntable; each first image in the first set of images includes at least one image marker of the feature surface; The at least one target camera is calibrated based on the first image set and a preset calibration algorithm to obtain the pose information of the at least one target camera relative to the target object; The at least one target camera is driven to capture images of a target object on the turntable based on the focusing parameters, thereby obtaining a second set of images of the target object, wherein the target object is placed at a predetermined position on the turntable; The target object is reconstructed in three dimensions based on the second image set and the pose information to generate a three-dimensional virtual model of the target object.

8. An image acquisition system, characterized in that, include: A turntable is used to place a target object or a calibration object, wherein the target object is an object that needs to be reconstructed in three dimensions; At least one target camera is used to photograph the target object or the calibration body placed on the turntable and to acquire image information of the target object or the calibration body; The calibration body includes: At least three feature surfaces, all of which are concave; the at least three feature surfaces are distributed circumferentially and connected in sequence; image markers are respectively provided on the at least three feature surfaces, so that when the target camera takes a picture of the calibration body, at least one image marker on the feature surface of the calibration body is within the imaging range of the target camera; The at least one target camera is further configured to focus on the target object on the turntable to obtain the focus parameters of the at least one target camera, wherein the target object is placed at a predetermined position on the turntable. The at least one target camera is further configured to take a picture of a calibration body on the turntable based on the focusing parameters, thereby obtaining a first image set of image identifiers on the calibration body, wherein the calibration body is placed at a predetermined position on the turntable; each first image in the first image set includes at least one image identifier of the feature surface; the at least one target camera is calibrated according to the first image set and a preset calibration algorithm to obtain the pose information of the at least one target camera relative to the target object.

9. An electronic device, characterized in that, include: At least one processor; as well as A memory that is communicatively connected to the at least one processor; The memory stores instructions executable by the at least one processor to cause the electronic device to perform the method of any one of claims 1-7.

10. A cloud device, characterized in that, include: At least one processor; as well as A memory that is communicatively connected to the at least one processor; The memory stores instructions that can be executed by the at least one processor, which, when executed by the at least one processor, cause the cloud device to perform the method according to any one of claims 1-7.

11. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, implement the method as described in any one of claims 1-7.

Images