Three-dimensional data acquisition method, device and equipment, computer readable medium and product

Abstract

The embodiment of the invention discloses a three-dimensional data acquisition method, device and equipment, a computer readable medium and a product. According to one specific embodiment, the method comprises the steps that in a preset time period, cameras in a camera set are controlled to synchronously collect part images of a target part of a target object, a part image group sequence is obtained, and the camera set comprises the cameras arranged at different visual angles of the target part; for each part image in the obtained part image group sequence, performing key point identification on the part image to obtain a two-dimensional coordinate set of a corresponding part key point set; and for each part image group in the part image group sequence, generating a three-dimensional coordinate set of the corresponding part key point set according to the camera parameter information set corresponding to the camera set and each obtained two-dimensional coordinate set of the corresponding part image group. According to the embodiment, the accuracy of the acquired three-dimensional data and the acquisition efficiency of the three-dimensional data are improved.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to three-dimensional data acquisition methods, devices, equipment, computer-readable media, and products. Background Art

[0002] 3D data acquisition refers to the use of a series of sensors or measuring devices to collect data on a three-dimensional object to be measured. At present, the commonly used methods for 3D data acquisition are: using a depth camera to collect 3D point cloud data to identify the 3D coordinates of the object to be measured; or using a binocular camera to identify the 3D coordinates of the object to be measured; or directly marking the coordinates of the object to be measured in a 3D coordinate system.

[0003] However, the inventors have found that when the above-mentioned method is used to collect three-dimensional data, the following technical problems often exist: when using a depth camera, the collected three-dimensional point cloud data is relatively one-sided, and the three-dimensional coordinates of other viewing angles of the object to be measured cannot be identified, resulting in poor accuracy of the collected three-dimensional data, and the cost of the depth camera is relatively high; when using a binocular camera, the three-dimensional coordinates of other viewing angles of the object to be measured cannot be identified, resulting in poor accuracy of the collected three-dimensional data; when using a three-dimensional coordinate system for data annotation, the annotation is difficult, resulting in low efficiency in three-dimensional data acquisition.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known to a person of ordinary skill in the art. Summary of the invention

[0005] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0006] Some embodiments of the present disclosure propose a three-dimensional data acquisition method, apparatus, electronic device, computer-readable medium, and computer program product to solve one or more of the technical problems mentioned in the above background technology section.

[0007] In a first aspect, some embodiments of the present disclosure provide a three-dimensional data acquisition method, the method comprising: within a preset time period, controlling each camera in a camera set to synchronously acquire a part image of a target part of a target object to obtain a part image group sequence, wherein each part image group in the above-mentioned part image group sequence is a part image acquired by the above-mentioned each camera at the same time, and the above-mentioned camera set includes each camera set at different viewing angles of the above-mentioned target part; for each part image in the obtained part image group sequence, performing key point recognition on the above-mentioned part image to obtain a two-dimensional coordinate set corresponding to the part key point set, wherein the above-mentioned part key point set corresponds to the above-mentioned target part; for each part image group in the above-mentioned part image group sequence, generating a three-dimensional coordinate set corresponding to the above-mentioned part key point set according to the camera parameter information set corresponding to the above-mentioned camera set and the obtained each two-dimensional coordinate set corresponding to the above-mentioned part image group.

[0008] Optionally, before performing key point identification on each part image in the obtained part image group sequence to obtain a two-dimensional coordinate set of the corresponding part key point set, the method further includes: performing image correction processing on each part image in the part image group sequence to update the part image group sequence.

[0009] Optionally, the method also includes: for the part image group in the above-mentioned part image group sequence that meets the conditions to be verified, performing part key point visualization processing on each part image in the above-mentioned part image group to obtain each part image after visualization processing, wherein each key point mark corresponding to the above-mentioned part key point set is displayed in each part image after visualization processing.

[0010] Optionally, the method also includes: in response to detecting a position adjustment operation of at least one key point mark corresponding to any part key point, determining the two-dimensional coordinate set corresponding to the part image group that meets the above-mentioned verification conditions as the target two-dimensional coordinate set; generating three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points according to each target two-dimensional coordinate corresponding to the above-mentioned arbitrary part key points in the above-mentioned target two-dimensional coordinate set and the adjusted two-dimensional coordinates of the above-mentioned at least one key point mark; and updating the three-dimensional coordinate set corresponding to the above-mentioned part image group according to the three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points.

[0011] Optionally, each part image after visualization also displays key point lines; and the method further comprises: updating each part image after visualization according to the three-dimensional coordinates corresponding to any of the key points of the above-mentioned parts.

[0012] Optionally, the above-mentioned performing part key point visualization processing on each part image in the above-mentioned part image group to obtain each part image after visualization processing includes: determining a three-dimensional coordinate set corresponding to the above-mentioned part image group; for each part image in the above-mentioned part image group, executing the following steps: determining each reprojected two-dimensional coordinate of the above-mentioned three-dimensional coordinate set corresponding to the above-mentioned part image to obtain a reprojected two-dimensional coordinate set; performing part key point marking processing in the above-mentioned part image according to the above-mentioned reprojected two-dimensional coordinate set to obtain a part image after marking processing, wherein each two-dimensional key point marking corresponding to the above-mentioned part key point set is displayed in the part image after marking processing; and displaying the part image after marking processing.

[0013] Optionally, the above-mentioned performing part key point visualization processing on each part image in the above-mentioned part image group to obtain each part image after visualization processing also includes: displaying a three-dimensional part model based on the determined three-dimensional coordinate set corresponding to the above-mentioned part image group, wherein each three-dimensional key point mark corresponding to the above-mentioned part key point set is displayed in the above-mentioned three-dimensional part model.

[0014] Optionally, the method further includes: in response to detecting a transformation operation on the visualized part image or the three-dimensional part model, transforming the visualized part image or the three-dimensional part model according to transformation mode information corresponding to the transformation operation.

[0015] Optionally, the above-mentioned generation of a three-dimensional coordinate set corresponding to the above-mentioned part key point set according to the camera parameter information set corresponding to the above-mentioned camera set and the obtained respective two-dimensional coordinate sets corresponding to the above-mentioned part image group includes: determining any two cameras in the above-mentioned camera set as an initial camera set; based on the initial camera set, performing the following first iteration step: for each part key point in the above-mentioned part key point set, performing the following steps: generating initial three-dimensional coordinates corresponding to the above-mentioned part key point according to the respective camera parameter information corresponding to the initial camera set in the above-mentioned camera parameter information set and the respective two-dimensional coordinate sets corresponding to the above-mentioned part image group; according to the above-mentioned part The invention relates to a method for generating key point reprojection error information corresponding to the key points of the above-mentioned parts based on the bit image group and the above-mentioned initial three-dimensional coordinates; generating reprojection error information according to the obtained reprojection error information of each key point; adding the generated reprojection error information to the reprojection error information set to update the reprojection error information set; in response to determining that the updated reprojection error information set satisfies the preset first iteration condition, selecting the reprojection error information that satisfies the preset first error condition from the updated reprojection error information set as the target reprojection error information; and determining each initial three-dimensional coordinate corresponding to the above-mentioned target reprojection error information as the three-dimensional coordinate set corresponding to the above-mentioned key point set.

[0016] Optionally, the above-mentioned iterative step also includes: in response to determining that the updated reprojection error information set does not satisfy the above-mentioned preset first iteration condition, adding an unused camera to the initial camera set to update the initial camera set, and based on the updated initial camera set, executing the above-mentioned first iteration step again.

[0017] Optionally, the above-mentioned generating key point reprojection error information corresponding to the above-mentioned part key points based on the above-mentioned part image group and the above-mentioned initial three-dimensional coordinates includes: determining each part image corresponding to the initial camera set in the above-mentioned part image group as the part image set to be measured; determining each two-dimensional coordinate corresponding to the above-mentioned part key points in each two-dimensional coordinate set corresponding to the above-mentioned part image set to be measured as the two-dimensional coordinate set to be measured; reprojecting the above-mentioned initial three-dimensional coordinates to each part image in the above-mentioned part image group to obtain each reprojected two-dimensional coordinate; generating key point reprojection error information corresponding to the above-mentioned part key points according to the above-mentioned each reprojected two-dimensional coordinate and the above-mentioned two-dimensional coordinate set to be measured.

[0018] Optionally, the above-mentioned three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points are generated according to the individual target two-dimensional coordinates corresponding to the above-mentioned arbitrary part key points in the above-mentioned target two-dimensional coordinate set and the adjusted two-dimensional coordinates of the above-mentioned at least one key point mark, including: determining the individual part images corresponding to the above-mentioned at least one key point mark in the part image group that meets the above-mentioned verification condition as a moving part image set; determining the above-mentioned moving part image set as an initial part image combination; determining the individual target two-dimensional coordinates corresponding to the above-mentioned arbitrary part key points in the above-mentioned target two-dimensional coordinate set as a key point two-dimensional coordinate set; based on the initial part image combination, performing the following second iterative step: determining the individual target two-dimensional coordinates corresponding to the above-mentioned arbitrary part key points in the above-mentioned key point two-dimensional coordinate set as a key point two-dimensional coordinate set. The marked two-dimensional coordinates are used to generate initial adjusted three-dimensional coordinates corresponding to the key points of any of the above-mentioned parts; according to the above-mentioned part image group and the above-mentioned initial adjusted three-dimensional coordinates, key point reprojection error information corresponding to the key points of any of the above-mentioned parts is generated; the above-mentioned key point reprojection error information is added to the key point reprojection error information set corresponding to the key points of any of the above-mentioned parts to update the key point reprojection error information set; in response to determining that the updated key point reprojection error information set satisfies the preset second iteration condition, the key point reprojection error information that satisfies the preset second error condition is selected from the updated key point reprojection error information set as the target key point reprojection error information; the initial adjusted three-dimensional coordinates corresponding to the above-mentioned target key point reprojection error information are determined as the three-dimensional coordinates corresponding to the key points of any of the above-mentioned parts.

[0019] Optionally, the above-mentioned second iteration step also includes: in response to determining that the updated key point reprojection error information set does not meet the above-mentioned preset second iteration condition, adding an unused part image in the part image group that meets the above-mentioned verification condition to the initial part image combination to update the initial part image combination, and based on the updated initial part image combination, executing the above-mentioned second iteration step again.

[0020] Optionally, the method also includes: for each part image in the above-mentioned part image group sequence, combining the reprojected two-dimensional coordinates and three-dimensional coordinates of each part key point in the part key point set corresponding to the above-mentioned part image into coordinate data to obtain a coordinate data set corresponding to the above-mentioned part image.

[0021] In a second aspect, some embodiments of the present disclosure provide a three-dimensional data acquisition device, the device comprising: an acquisition unit, configured to control each camera in a camera set to synchronously acquire a part image of a target part of a target object within a preset time period, to obtain a part image group sequence, wherein each part image group in the above-mentioned part image group sequence is a part image acquired by the above-mentioned each camera at the same time, and the above-mentioned camera set comprises each camera set at different viewing angles of the above-mentioned target part; an identification unit, configured to perform key point identification on each part image in the obtained part image group sequence, to obtain a two-dimensional coordinate set corresponding to the part key point set, wherein the above-mentioned part key point set corresponds to the above-mentioned target part; a generation unit, configured to generate a three-dimensional coordinate set corresponding to the above-mentioned part key point set for each part image group in the above-mentioned part image group sequence according to a camera parameter information set corresponding to the above-mentioned camera set and each obtained two-dimensional coordinate set corresponding to the above-mentioned part image group.

[0022] Optionally, before the recognition unit, the three-dimensional data acquisition device further includes: a correction unit configured to perform image correction processing on each part image in the part image group sequence to update the part image group sequence.

[0023] Optionally, the three-dimensional data acquisition device also includes: a visualization unit, which is configured to perform part key point visualization processing on each part image in the part image group that meets the verification conditions in the above-mentioned part image group sequence, so as to obtain each part image after visualization processing, wherein each key point mark corresponding to the above-mentioned part key point set is displayed in each part image after visualization processing.

[0024] Optionally, the three-dimensional data acquisition device further includes: a determination unit, a coordinate generation unit and an update unit. The determination unit is configured to determine the two-dimensional coordinate set corresponding to the part image group that meets the above-mentioned verification condition as the target two-dimensional coordinate set in response to detecting the position adjustment operation of at least one key point mark corresponding to the key point of any part. The coordinate generation unit is configured to generate the three-dimensional coordinates corresponding to the above-mentioned key point of any part according to the target two-dimensional coordinates corresponding to the above-mentioned key point of any part in the above-mentioned target two-dimensional coordinate set and the adjusted two-dimensional coordinates of the above-mentioned at least one key point mark. The update unit is configured to update the three-dimensional coordinate set corresponding to the above-mentioned part image group according to the three-dimensional coordinates corresponding to the above-mentioned key point of any part.

[0025] Optionally, key point lines are also displayed in each part image after visualization processing.

[0026] Optionally, the three-dimensional data acquisition device further includes: an image updating unit configured to update the images of each part after visualization processing according to the three-dimensional coordinates corresponding to the key points of any of the above parts.

[0027] Optionally, the visualization unit is further configured to: determine a three-dimensional coordinate set corresponding to the above-mentioned part image group; for each part image in the above-mentioned part image group, perform the following steps: determine each reprojected two-dimensional coordinate of the above-mentioned part image corresponding to the above-mentioned three-dimensional coordinate set to obtain a reprojected two-dimensional coordinate set; perform part key point marking processing in the above-mentioned part image according to the above-mentioned reprojected two-dimensional coordinate set to obtain a marked part image, wherein each two-dimensional key point marking corresponding to the above-mentioned part key point set is displayed in the marked part image; and display the marked part image.

[0028] Optionally, the visualization unit is further configured to: display a three-dimensional part model according to the determined three-dimensional coordinate set corresponding to the above-mentioned part image group, wherein the above-mentioned three-dimensional part model displays each three-dimensional key point mark corresponding to the above-mentioned part key point set.

[0029] Optionally, the three-dimensional data acquisition device also includes: a transformation unit, which is configured to respond to detecting a transformation operation on the visualized part image or the above-mentioned three-dimensional part model, and transform the above-mentioned part image or the above-mentioned three-dimensional part model according to transformation mode information corresponding to the above-mentioned transformation operation.

[0030] Optionally, the generating unit is further configured to: determine any two cameras in the above-mentioned camera set as the initial camera set; based on the initial camera set, perform the following first iteration step: for each part key point in the above-mentioned part key point set, perform the following steps: generate initial three-dimensional coordinates corresponding to the above-mentioned part key points according to each camera parameter information corresponding to the initial camera set in the above-mentioned camera parameter information set and each two-dimensional coordinate set corresponding to the above-mentioned part image group; generate key point reprojection error information corresponding to the above-mentioned part key points according to the above-mentioned part image group and the above-mentioned initial three-dimensional coordinates; generate reprojection error information according to the obtained reprojection error information of each key point; add the generated reprojection error information to the reprojection error information set to update the reprojection error information set; in response to determining that the updated reprojection error information set satisfies the preset first iteration condition, select the reprojection error information that satisfies the preset first error condition from the updated reprojection error information set as the target reprojection error information; determine each initial three-dimensional coordinate corresponding to the above-mentioned target reprojection error information as the three-dimensional coordinate set corresponding to the above-mentioned part key point set.

[0031] Optionally, the above-mentioned iterative step also includes: in response to determining that the updated reprojection error information set does not satisfy the above-mentioned preset first iteration condition, adding an unused camera to the initial camera set to update the initial camera set, and based on the updated initial camera set, executing the above-mentioned first iteration step again.

[0032] Optionally, the generating unit is further configured to: determine each part image corresponding to the initial camera set in the above-mentioned part image group as the part image set to be measured; determine each two-dimensional coordinate corresponding to the above-mentioned part key point in each two-dimensional coordinate set corresponding to the above-mentioned part image set to be measured as the two-dimensional coordinate set to be measured; reproject the above-mentioned initial three-dimensional coordinates to each part image in the above-mentioned part image group to obtain each reprojected two-dimensional coordinate; generate key point reprojection error information corresponding to the above-mentioned part key point according to the above-mentioned each reprojected two-dimensional coordinate and the above-mentioned two-dimensional coordinate set to be measured.

[0033] Optionally, the coordinate generating unit is further configured to: determine the various part images corresponding to the at least one key point mark in the part image group that meets the above-mentioned verification condition as a moving part image set; determine the above-mentioned moving part image set as an initial part image combination; determine the various target two-dimensional coordinates corresponding to the above-mentioned arbitrary part key points in the above-mentioned target two-dimensional coordinate set as a key point two-dimensional coordinate set; based on the initial part image combination, perform the following second iteration step: generate initial adjusted three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points according to the various target two-dimensional coordinates in the above-mentioned key point two-dimensional coordinate set corresponding to the initial part image combination and the adjusted two-dimensional coordinates of the at least one key point mark; based on According to the above-mentioned part image group and the above-mentioned initial adjusted three-dimensional coordinates, key point reprojection error information corresponding to the above-mentioned arbitrary part key points is generated; the above-mentioned key point reprojection error information is added to the key point reprojection error information set corresponding to the above-mentioned arbitrary part key points to update the key point reprojection error information set; in response to determining that the updated key point reprojection error information set satisfies the preset second error condition, the key point reprojection error information that satisfies the preset second error condition is selected from the updated key point reprojection error information set as the target key point reprojection error information; the initial adjusted three-dimensional coordinates corresponding to the above-mentioned target key point reprojection error information are determined as the three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points.

[0034] Optionally, the above-mentioned second iteration step also includes: in response to determining that the updated key point reprojection error information set does not meet the above-mentioned preset second iteration condition, adding an unused part image in the part image group that meets the above-mentioned verification condition to the initial part image combination to update the initial part image combination, and based on the updated initial part image combination, executing the above-mentioned second iteration step again.

[0035] Optionally, the three-dimensional data acquisition device also includes: a combination unit, which is configured to combine the reprojected two-dimensional coordinates and three-dimensional coordinates of each part key point in the part key point set corresponding to the above-mentioned part image into coordinate data for each part image in the above-mentioned part image group sequence, so as to obtain a coordinate data set corresponding to the above-mentioned part image.

[0036] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner of the above-mentioned first aspect.

[0037] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.

[0038] In a fifth aspect, some embodiments of the present disclosure provide a computer program product, including a computer program, which implements the method described in any implementation manner of the above-mentioned first aspect when executed by a processor.

[0039] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the three-dimensional data acquisition method of some embodiments of the present disclosure, the accuracy of the collected three-dimensional data and the acquisition efficiency of the three-dimensional data are improved. Specifically, the reasons for the poor accuracy and low acquisition efficiency of the collected three-dimensional data are: using a depth camera, the collected three-dimensional point cloud data is one-sided, and the three-dimensional coordinates of other viewing angles of the object to be measured cannot be identified, resulting in poor accuracy of the collected three-dimensional data, and the cost of the depth camera is high; using a binocular camera, the three-dimensional coordinates of other viewing angles of the object to be measured cannot be identified, resulting in poor accuracy of the collected three-dimensional data; using a three-dimensional coordinate system for data annotation, the annotation difficulty is large, resulting in low efficiency of three-dimensional data acquisition. Based on this, the three-dimensional data acquisition method of some embodiments of the present disclosure, first, within a preset time period, control each camera in the camera set to synchronously acquire the part image of the target part of the target object, and obtain a part image group sequence, wherein each part image group in the above-mentioned part image group sequence is a part image acquired by each camera at the same time, and the above-mentioned camera set includes each camera set at different viewing angles of the above-mentioned target part. Thus, multiple frames of images of the target part can be synchronously acquired by the camera in front of multiple viewing angles of the target object. Then, for each part image in the obtained part image group sequence, the above part image is subjected to key point recognition to obtain a two-dimensional coordinate set of the corresponding part key point set, wherein the above part key point set corresponds to the above target part. Thus, the two-dimensional coordinates of each key point of the target part in each image can be pre-identified. Finally, for each part image group in the above part image group sequence, a three-dimensional coordinate set corresponding to the above part key point set is generated according to the camera parameter information set corresponding to the above camera set and the obtained two-dimensional coordinate sets corresponding to the above part image group. Thus, the three-dimensional coordinates of each key point can be directly determined by each two-dimensional coordinate of each key point of the target part under multiple viewing angles. Also, because the two-dimensional images are collected by each camera in front of the target object at multiple viewing angles, it is not necessary to use a depth camera, and the comprehensiveness of the collected three-dimensional data can be improved. Also, because the three-dimensional coordinates of each key point can be directly determined by each two-dimensional coordinate of each key point, there is no need to manually mark the three-dimensional coordinates, which improves the collection efficiency of the three-dimensional data. Thus, the accuracy of the collected three-dimensional data and the collection efficiency of the three-dimensional data are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0041] Figure 1 is a flow chart of some embodiments of the three-dimensional data acquisition method according to the present disclosure;

[0042] Figure 2 are flow charts of other embodiments of the three-dimensional data acquisition method according to the present disclosure;

[0043] Figure 3 is a schematic structural diagram of some embodiments of the three-dimensional data acquisition device according to the present disclosure;

[0044] Figure 4 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0045] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0046] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0047] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0048] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0049] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0050] With regard to the collection, storage, and use of user personal information (such as body images) involved in this disclosure, before performing the corresponding operations, the relevant organizations or individuals shall fulfill their obligations, including conducting personal information security impact assessments, fulfilling the obligation to inform the personal information subject, and obtaining the authorization and consent of the personal information subject in advance.

[0051] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0052] Figure 1 The process 100 of some embodiments of the three-dimensional data acquisition method according to the present disclosure is shown. The three-dimensional data acquisition method comprises the following steps:

[0053] Step 101 : within a preset time period, control each camera in a camera set to synchronously capture a part image of a target part of a target object to obtain a part image group sequence.

[0054] In some embodiments, the execution subject of the three-dimensional data acquisition method (such as the computing device shown) can control each camera in the camera set to synchronously acquire the part image of the target part of the target object through a wired connection or a wireless connection within a preset time period to obtain a part image group sequence. The camera set includes each camera set at different viewing angles of the target part. The camera set may include at least 3 cameras. The camera set may be set at different positions of the bracket to acquire images of the target part of the target object from different viewing angles. Each camera in the camera set may achieve at least one of the following: synchronous image acquisition, global exposure, higher frame rate, and acquisition of images with the same resolution. The placement position of each camera may cover as many angles of the hand as possible. The viewing angle of each camera can be adjusted to cover as much as possible the range of movement of the target part during acquisition. The internal parameter information of each camera may be pre-calibrated. The internal parameter information may include but is not limited to: focal length, principal point coordinates. The external parameter information corresponding to each camera may also be pre-calibrated. The external parameter information may include but is not limited to: rotation matrix, translation vector. The execution subject and each camera may be connected in communication, and the execution subject may control each camera to capture images. For example, the execution subject may be a notebook. The execution body may also be placed on the bracket. Optionally, the bracket may further include a shoulder strap for the user to carry and move in a back-mounted manner. The target object may be an object for which three-dimensional data is to be collected. The target part may be a part of the target object. For example, the target object may be a user for whom three-dimensional data is to be collected. The target part may be a hand of the user. For the convenience of description, in the embodiments of the present disclosure, one target part is described. In practice, the same three-dimensional data collection method may be used for multiple target parts. The part image group sequence may be a group of part images continuously collected by each camera within the preset time period. Each part image group corresponds to a collection time point. The part images in the part image group may be images of the target part collected synchronously by each camera at the collection time point.

[0055] It should be noted that the above-mentioned wireless connection methods may include but are not limited to 3G / 4G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (ultra wideband) connection, and other wireless connection methods currently known or to be developed in the future. The above-mentioned bracket may be made of lightweight materials. For example, lightweight materials may be plastic or composite materials instead of steel to reduce the weight of the equipment and facilitate user movement. The above-mentioned bracket can be directly folded or adjusted in height to adapt to different users and collection environments, and image data with different backgrounds, different lighting, and different viewing angles can be collected by moving the bracket position. Brackets of different sizes can be configured for target objects of different sizes.

[0056] Optionally, before step 102, the execution subject may perform image correction processing on each part image in the part image group sequence to update the part image group sequence. In practice, the execution subject may perform image distortion correction processing on each part image in the part image group sequence by using the internal reference information of each camera and the external reference information corresponding to each camera to update the part image group sequence. Thus, image deformation can be eliminated before performing two-dimensional key point detection.

[0057] Step 102 : for each part image in the obtained part image group sequence, key point recognition is performed on the part image to obtain a two-dimensional coordinate set of a corresponding part key point set.

[0058] In some embodiments, for each part image in the obtained part image group sequence, the above-mentioned execution subject may perform key point recognition on the above-mentioned part image to obtain a two-dimensional coordinate set of the corresponding part key point set. The above-mentioned part key point set corresponds to the above-mentioned target part. The above-mentioned part key point set may be each pre-set structural point that can characterize the key structural point position of the above-mentioned target part. For example, when the target part is a hand, the key point set may include 21 main bone nodes. In the two-dimensional coordinate set, each two-dimensional coordinate may correspond to a part key point. In practice, the above-mentioned execution subject may input the above-mentioned part image into a part key point recognition model corresponding to the above-mentioned target part that is pre-trained to obtain a two-dimensional coordinate set of the corresponding part key point set. The part key point recognition model may be a neural network model that takes the part image as input and takes the two-dimensional coordinate set of the corresponding part key point set as output. For example, when the target part is a hand, the above-mentioned part key point recognition model may be a hand key point recognition model. The hand key point recognition model may be an OpenPose model.

[0059] Step 103 , for each part image group in the part image group sequence, generate a three-dimensional coordinate set of the corresponding part key point set according to the camera parameter information set corresponding to the camera set and each obtained two-dimensional coordinate set of the corresponding part image group.

[0060] In some embodiments, for each part image group in the above-mentioned part image group sequence, the above-mentioned execution subject may generate a three-dimensional coordinate set corresponding to the above-mentioned part key point set according to the camera parameter information set corresponding to the above-mentioned camera set and the obtained respective two-dimensional coordinate sets corresponding to the above-mentioned part image group. The above-mentioned camera parameter information set may be camera parameter related information corresponding to each camera. Each camera parameter information may correspond to a camera. The camera parameter information may include internal reference information and may also include external reference information. In practice, for each part key point in the above-mentioned part key point set, first, the above-mentioned execution subject may determine the respective two-dimensional coordinates corresponding to the above-mentioned part key point in the above-mentioned part image group as a part two-dimensional coordinate set. Then, the three-dimensional coordinates corresponding to the above-mentioned part key point may be generated according to the above-mentioned part two-dimensional coordinate set and the above-mentioned camera parameter information set. Specifically, the above-mentioned execution subject may generate the three-dimensional coordinates corresponding to the above-mentioned part key point according to the above-mentioned part two-dimensional coordinate set and the above-mentioned camera parameter information set through a triangulation algorithm. The triangulation algorithm may be a multi-eye triangulation method.

[0061] In some optional implementations of some embodiments, the execution subject may generate a three-dimensional coordinate set corresponding to the above-mentioned part key point set according to the camera parameter information set corresponding to the above-mentioned camera set and each obtained two-dimensional coordinate set corresponding to the above-mentioned part image group through the following steps:

[0062] In the first step, any two cameras in the above camera set are determined as the initial camera set.

[0063] In the second step, based on the initial camera set, perform the following first iteration steps:

[0064] In the first sub-step, for each part key point in the above part key point set, perform the following steps:

[0065] First, based on the camera parameter information corresponding to the initial camera set in the camera parameter information set and the two-dimensional coordinate sets corresponding to the part image group, the initial three-dimensional coordinates corresponding to the key points of the part are generated. In practice, the execution subject may first select the two-dimensional coordinate sets corresponding to the initial camera set from the two-dimensional coordinate sets corresponding to the part image group. Then, the two-dimensional coordinates corresponding to the key points of the part in the selected two-dimensional coordinate sets may be determined as the part two-dimensional coordinate set. Then, based on the part two-dimensional coordinate set and the camera parameter information, the three-dimensional coordinates corresponding to the key points of the part may be generated as the initial three-dimensional coordinates. Specifically, the execution subject may generate the three-dimensional coordinates corresponding to the key points of the part by a triangulation algorithm.

[0066] Second, based on the above-mentioned part image group and the above-mentioned initial three-dimensional coordinates, generate key point reprojection error information corresponding to the above-mentioned part key points. In practice, the above-mentioned execution entity may first reproject the above-mentioned initial three-dimensional coordinates to each part image in the above-mentioned part image group according to each camera parameter information of the corresponding initial camera set to obtain each reprojected two-dimensional coordinate. Then, for each reprojected two-dimensional coordinate obtained, the above-mentioned execution entity may determine the distance between the above-mentioned reprojected two-dimensional coordinate and the two-dimensional coordinate corresponding to the above-mentioned reprojected two-dimensional coordinate in the corresponding part image. Finally, the sum of each obtained distance may be determined as the key point reprojection error information corresponding to the above-mentioned part key points.

[0067] The second sub-step is to generate reprojection error information according to the obtained reprojection error information of each key point. In practice, the above execution subject may determine the sum of the obtained reprojection error information of each key point as the reprojection error information.

[0068] The third sub-step is to add the generated reprojection error information to the reprojection error information set to update the reprojection error information set. When the first iteration step is performed for the first time, the reprojection error information set may be empty.

[0069] In a fourth sub-step, in response to determining that the updated reprojection error information set satisfies the preset first iteration condition, reprojection error information satisfying the preset first error condition is selected from the updated reprojection error information set as the target reprojection error information. The preset first iteration condition may be that any combination of two or more cameras in the camera set is used as the initial camera set during the iteration process. The preset first iteration condition may also be that the number of executions of the first iteration step is greater than or equal to the preset number of times. The preset first iteration condition may also be that at least one preset combination of two or more cameras in the camera set is used as the initial camera set. The above-mentioned preset first error condition may be that the reprojection error information is minimized.

[0070] The fifth sub-step is to determine each initial three-dimensional coordinate corresponding to the target reprojection error information as a three-dimensional coordinate set corresponding to the key point set of the part.

[0071] In a sixth sub-step, in response to determining that the updated reprojection error information set does not satisfy the above-mentioned preset first iteration condition, an unused camera is added to the initial camera set to update the initial camera set, and the above-mentioned first iteration step is performed again based on the updated initial camera set. Thus, the three-dimensional coordinates with the smallest error after reprojection can be determined by adaptive iterative multi-eye triangulation, which can further improve the accuracy of the three-dimensional data.

[0072] Optionally, the execution subject may generate key point reprojection error information corresponding to the key points of the part according to the part image group and the initial three-dimensional coordinates through the following steps:

[0073] In the first step, each part image corresponding to the initial camera set in the part image group is determined as the part image set to be tested.

[0074] In the second step, each two-dimensional coordinate corresponding to the key point of the above-mentioned part in each two-dimensional coordinate set corresponding to the above-mentioned image set of the part to be measured is determined as the two-dimensional coordinate set to be measured.

[0075] The third step is to reproject the initial three-dimensional coordinates onto each part image in the part image group to obtain each reprojected two-dimensional coordinate.

[0076] The fourth step is to generate the key point reprojection error information corresponding to the key point of the above-mentioned part according to the above-mentioned reprojected two-dimensional coordinates and the above-mentioned two-dimensional coordinate set to be measured. In practice, for each of the above-mentioned reprojected two-dimensional coordinates, the execution entity can determine the distance between the above-mentioned reprojected two-dimensional coordinate and the two-dimensional coordinate to be measured corresponding to the above-mentioned reprojected two-dimensional coordinate in the above-mentioned two-dimensional coordinate set to be measured as the coordinate distance. Then, the sum of the determined coordinate distances can be determined as the key point reprojection error information. Therefore, the key point reprojection error information corresponding to a key point of a part can be determined by using the distance between the two-dimensional coordinate obtained after reprojection in each part image and the original two-dimensional coordinate.

[0077] Optionally, for each part image in the part image group sequence, the execution subject may further combine the reprojected two-dimensional coordinates and three-dimensional coordinates of each part key point in the part key point set corresponding to the part image into coordinate data to obtain a coordinate data set corresponding to the part image. Thus, the reprojected two-dimensional coordinates and three-dimensional coordinates may be stored in pairs to facilitate retrieval of the reprojected two-dimensional coordinates and three-dimensional coordinates.

[0078] Optionally, when the target part is a hand, the execution subject may also determine the angle of the finger line on the obtained three-dimensional coordinates of the key points of each part, so as to filter out or correct some abnormal data that violates the finger bending rules.

[0079] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the three-dimensional data acquisition method of some embodiments of the present disclosure, the accuracy of the collected three-dimensional data and the acquisition efficiency of the three-dimensional data are improved. Specifically, the reasons for the poor accuracy and low acquisition efficiency of the collected three-dimensional data are: using a depth camera, the collected three-dimensional point cloud data is one-sided, and the three-dimensional coordinates of other viewing angles of the object to be measured cannot be identified, resulting in poor accuracy of the collected three-dimensional data, and the cost of the depth camera is high; using a binocular camera, the three-dimensional coordinates of other viewing angles of the object to be measured cannot be identified, resulting in poor accuracy of the collected three-dimensional data; using a three-dimensional coordinate system for data annotation, the annotation difficulty is large, resulting in low efficiency of three-dimensional data acquisition. Based on this, the three-dimensional data acquisition method of some embodiments of the present disclosure, first, within a preset time period, control each camera in the camera set to synchronously acquire the part image of the target part of the target object, and obtain a part image group sequence, wherein each part image group in the above-mentioned part image group sequence is a part image acquired by each camera at the same time, and the above-mentioned camera set includes each camera set at different viewing angles of the above-mentioned target part. Thus, multiple frames of images of the target part can be synchronously acquired by the camera in front of multiple viewing angles of the target object. Then, for each part image in the obtained part image group sequence, the above part image is subjected to key point recognition to obtain a two-dimensional coordinate set of the corresponding part key point set, wherein the above part key point set corresponds to the above target part. Thus, the two-dimensional coordinates of each key point of the target part in each image can be pre-identified. Finally, for each part image group in the above part image group sequence, a three-dimensional coordinate set corresponding to the above part key point set is generated according to the camera parameter information set corresponding to the above camera set and the obtained two-dimensional coordinate sets corresponding to the above part image group. Thus, the three-dimensional coordinates of each key point can be directly determined by each two-dimensional coordinate of each key point of the target part under multiple viewing angles. Also, because the two-dimensional images are collected by each camera in front of the target object at multiple viewing angles, it is not necessary to use a depth camera, and the comprehensiveness of the collected three-dimensional data can be improved. Also, because the three-dimensional coordinates of each key point can be directly determined by each two-dimensional coordinate of each key point, there is no need to manually mark the three-dimensional coordinates, which improves the collection efficiency of the three-dimensional data. Thus, the accuracy of the collected three-dimensional data and the collection efficiency of the three-dimensional data are improved.

[0080] Further references Figure 2 , which shows a process 200 of another embodiment of a three-dimensional data acquisition method. The process 200 of the three-dimensional data acquisition method includes the following steps:

[0081] Step 201 : within a preset time period, control each camera in a camera set to synchronously capture a part image of a target part of a target object to obtain a part image group sequence.

[0082] Step 202 : for each part image in the obtained part image group sequence, key point recognition is performed on the part image to obtain a two-dimensional coordinate set of a corresponding part key point set.

[0083] Step 203 , for each part image group in the part image group sequence, generate a three-dimensional coordinate set of the corresponding part key point set according to the camera parameter information set corresponding to the camera set and each obtained two-dimensional coordinate set of the corresponding part image group.

[0084] In some embodiments, the specific implementation of steps 201-203 and the technical effects brought about can be referred to Figure 1 The corresponding steps 101-103 in the embodiments are not described in detail here.

[0085] Step 204 , for the part image groups in the part image group sequence that meet the to-be-verified condition, perform part key point visualization processing on each part image in the part image group to obtain each part image after visualization processing.

[0086] In some embodiments, for the part image group that meets the condition to be verified in the above-mentioned part image group sequence, the execution subject of the three-dimensional data acquisition method can perform part key point visualization processing on each part image in the above-mentioned part image group to obtain each part image after visualization processing. Among them, the above-mentioned condition to be verified can be that the part image group is the first part image group in the part image group sequence. The above-mentioned condition to be verified can also be that the part image group is a part image group selected by the user. The user can select the part image group in the interface, or switch the selected part image group between various part image groups by switching controls or keys on the keyboard. Each part image after visualization processing is displayed with each key point mark corresponding to the above-mentioned part key point set. In practice, the above-mentioned execution subject can display each part image in the part image group in the display interface, and display each key point mark corresponding to the part image in each displayed part image. The display position of each key point mark is each two-dimensional coordinate after reprojection according to the three-dimensional coordinate set corresponding to the part key point set.

[0087] In some optional implementations of some embodiments, the execution subject may perform part key point visualization processing on each part image in the part image group to obtain each part image after visualization processing through the following steps:

[0088] The first step is to determine the three-dimensional coordinate set corresponding to the above-mentioned part image group.

[0089] In the second step, for each part image in the above part image group, the following steps are performed:

[0090] The first sub-step is to determine the reprojected two-dimensional coordinates of the three-dimensional coordinate set corresponding to the part image to obtain a reprojected two-dimensional coordinate set. The reprojected two-dimensional coordinates may be the two-dimensional coordinates of the part image reprojected from the three-dimensional coordinates.

[0091] The second sub-step is to perform a part key point marking process in the part image according to the above-mentioned reprojected two-dimensional coordinate set to obtain the part image after the marking process. The part image after the marking process displays the respective two-dimensional key point marks corresponding to the above-mentioned part key point set. In practice, the above-mentioned execution subject can add two-dimensional key point marks to each position corresponding to the above-mentioned reprojected two-dimensional coordinate set in the part image. For example, the two-dimensional key point mark can be a dot mark. There is no specific limitation on the representation form of the two-dimensional key point mark.

[0092] The third sub-step is to display the marked part image.

[0093] Optionally, the above-mentioned execution subject can also display a three-dimensional part model according to the determined three-dimensional coordinate set corresponding to the above-mentioned part image group. Among them, the above-mentioned three-dimensional part model displays the three-dimensional key point marks corresponding to the above-mentioned part key point set. The above-mentioned three-dimensional part model can be a three-dimensional modeling model of the above-mentioned target part. In practice, the above-mentioned execution subject can display the three-dimensional part model constructed by the above-mentioned three-dimensional coordinate set in a three-dimensional coordinate system. The three-dimensional key point mark can be represented in the form of a dot mark or a spherical mark. There is no specific limitation on the representation form of the three-dimensional key point mark. The three-dimensional part model can be displayed together with the various part images after visualization processing. For example, the three-dimensional part model can be displayed on the right side of the various part images after visualization processing, so that the user can verify the visualized two-dimensional data and three-dimensional data.

[0094] Step 205 , in response to detecting a position adjustment operation of at least one key point mark corresponding to any part key point, determining a two-dimensional coordinate set corresponding to the part image group that meets the verification condition as a target two-dimensional coordinate set.

[0095] In some embodiments, the execution subject may determine the two-dimensional coordinate set corresponding to the group of images of the part that meets the conditions to be verified as the target two-dimensional coordinate set in response to detecting a position adjustment operation of at least one key point mark corresponding to any part key point. The position adjustment operation may be an operation of moving the position of the key point mark in the displayed part image. Optionally, after detecting the position adjustment operation of at least one key point mark corresponding to any part key point, the execution subject may determine the two-dimensional coordinate set corresponding to the group of images of the part that meets the conditions to be verified as the target two-dimensional coordinate set in response to detecting a selection operation acting on a confirmation and verification control. Thus, the three-dimensional coordinates of the above-mentioned key points of any part may be intelligently verified through the confirmation and verification control according to the position of at least one key point mark that has been adjusted.

[0096] Step 206 , generating three-dimensional coordinates corresponding to the key points of any part according to the target two-dimensional coordinates corresponding to the key points of any part in the target two-dimensional coordinate set and the adjusted two-dimensional coordinates of at least one key point mark.

[0097] In some embodiments, the execution entity may generate three-dimensional coordinates corresponding to the key points of any part according to the target two-dimensional coordinates corresponding to the key points of any part in the target two-dimensional coordinate set and the adjusted two-dimensional coordinates of at least one key point mark.

[0098] In some optional implementations of some embodiments, the execution subject may generate the three-dimensional coordinates corresponding to the key points of the arbitrary parts according to the target two-dimensional coordinates corresponding to the key points of the arbitrary parts in the target two-dimensional coordinate set and the adjusted two-dimensional coordinates of the at least one key point mark through the following steps:

[0099] In the first step, each part image corresponding to the at least one key point mark in the part image group satisfying the above-mentioned verification condition is determined as a moving part image set.

[0100] In the second step, the above moving part image set is determined as the initial part image combination. The initial part image combination can be used as a combination variable in the second iteration step.

[0101] The third step is to determine each target two-dimensional coordinate corresponding to the above arbitrary part key point in the above target two-dimensional coordinate set as a key point two-dimensional coordinate set. Thus, the key point two-dimensional coordinate set can be used as the two-dimensional coordinate of the above arbitrary part key point before moving in each image.

[0102] The fourth step is to perform the following second iteration steps based on the initial part image combination:

[0103] The first sub-step is to generate the initial adjusted three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points according to the target two-dimensional coordinates corresponding to the initial part image combination in the above-mentioned key point two-dimensional coordinate set and the adjusted two-dimensional coordinates of the above-mentioned at least one key point mark. In practice, the above-mentioned execution subject can determine the target two-dimensional coordinates corresponding to the initial part image combination in the above-mentioned key point two-dimensional coordinate set as the first target two-dimensional coordinate set. Then, the first target two-dimensional coordinates that meet the unadjusted condition in the above-mentioned first target two-dimensional coordinate set can be determined as the second target two-dimensional coordinate set. After that, the three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points can be generated as the initial adjusted three-dimensional coordinates according to the above-mentioned second target two-dimensional coordinate set, the adjusted two-dimensional coordinates of the above-mentioned at least one key point mark and the various camera parameter information corresponding to the above-mentioned initial part image combination. Specifically, the above-mentioned execution subject can generate the three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points through a triangulation algorithm.

[0104] The second sub-step is to generate the key point reprojection error information corresponding to any of the above-mentioned key points of the part according to the above-mentioned part image group and the above-mentioned initial adjusted three-dimensional coordinates. In practice, the above-mentioned execution subject may first reproject the above-mentioned initial adjusted three-dimensional coordinates to each part image in the above-mentioned part image group according to each camera parameter information corresponding to the initial part image combination, and obtain each reprojected two-dimensional coordinate. Then, for each reprojected two-dimensional coordinate obtained, the above-mentioned execution subject may determine the distance between the above-mentioned reprojected two-dimensional coordinate and the two-dimensional coordinate corresponding to the above-mentioned reprojected two-dimensional coordinate in the corresponding part image. Finally, the sum of each obtained distance may be determined as the key point reprojection error information corresponding to the above-mentioned key point of the part.

[0105] The third sub-step is to add the key point reprojection error information to the key point reprojection error information set corresponding to the key point of any part to update the key point reprojection error information set. When the second iteration step is performed for the first time, the key point reprojection error information set may be empty.

[0106] In a fourth sub-step, in response to determining that the updated key point reprojection error information set satisfies the preset second iteration condition, the key point reprojection error information that satisfies the preset second error condition is selected from the updated key point reprojection error information set as the target key point reprojection error information. The preset second iteration condition may be that any combination of the part image group that is greater than or equal to the above-mentioned moving part image set is used as the initial part image combination during the iteration process. The preset second iteration condition may also be that the number of executions of the second iteration step is greater than or equal to the preset number of times. The above-mentioned preset second error condition may be that the reprojection error information is minimum.

[0107] In a fifth sub-step, the initial adjusted three-dimensional coordinates corresponding to the reprojection error information of the target key points are determined as the three-dimensional coordinates corresponding to the key points of any of the above parts.

[0108] The sixth sub-step is, in response to determining that the updated key point reprojection error information set does not meet the above-mentioned preset second iteration condition, adding an unused part image in the part image group that meets the above-mentioned condition to be verified to the initial part image combination to update the initial part image combination, and performing the above-mentioned second iteration step again based on the updated initial part image combination. Thus, the three-dimensional coordinates of the adjusted part key points can be calibrated by determining the three-dimensional coordinates with the smallest error after reprojection through the adaptive iterative multi-eye triangulation method, so as to further improve the accuracy of the calibrated three-dimensional data.

[0109] Step 207: Update the three-dimensional coordinate set corresponding to the part image group according to the three-dimensional coordinates of the key points corresponding to any part.

[0110] In some embodiments, the execution subject may update the three-dimensional coordinate set corresponding to the above-mentioned part image group according to the three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points. In practice, the execution subject may replace the three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points in the above-mentioned three-dimensional coordinate set with the currently generated three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points, so as to update the three-dimensional coordinate set corresponding to the above-mentioned part image group.

[0111] Optionally, key point lines can also be displayed in each part image after visualization processing. The key point lines can be lines between key points to characterize the structure of the target part. For example, when the target part is the hand, the key point lines can be lines from a point on the back of the hand to the fingertips of each finger. The key point lines corresponding to different fingers can be displayed in different colors. The above-mentioned execution entity can also update the images of each part after visualization processing according to the three-dimensional coordinates corresponding to the key points of any of the above-mentioned parts. In practice, the above-mentioned execution entity can reproject the above-mentioned three-dimensional coordinates to the above-mentioned images of each part to obtain each reprojected two-dimensional coordinate. Then, the position of the key point mark of the key point of any of the above-mentioned parts can be updated in each part image according to each reprojected two-dimensional coordinate.

[0112] Optionally, the execution subject may also transform the visualized part image or the three-dimensional part model in response to detecting a transformation operation on the visualized part image or the three-dimensional part model, according to the transformation mode information corresponding to the transformation operation. The transformation operation may include but is not limited to at least one of the following operations: move, rotate, zoom in, and zoom out. The transformation mode information may include a transformation type and a transformation parameter. For example, when the transformation operation is a move operation, the transformation type may be move, and the transformation parameter may include a moving distance and a moving direction. When the transformation operation is a rotation operation, the transformation type may be rotation, and the transformation parameter may include a rotation angle. When the transformation operation is a zoom-in operation, the transformation type may be zoom-in, and the transformation parameter may include a zoom-in multiple. When the transformation operation is a zoom-out operation, the transformation type may be zoom-out, and the transformation parameter may include a zoom-out multiple. Thus, during verification, an image or model transformation path may be provided to the user for easy viewing.

[0113] Optionally, a reset control may also be displayed in the display interface. The execution subject may, in response to detecting a selection operation acting on the reset control, reset the key point marks displayed in the displayed part image according to the initial positions of the key point marks. The three-dimensional part model may also be reset accordingly.

[0114] from Figure 2 It can be seen that Figure 1 Compared with the description of some corresponding embodiments, Figure 2 The process 200 of the three-dimensional data acquisition method in some corresponding embodiments embodies the steps expanded by visualizing the key point markings. Therefore, the schemes described in these embodiments can use the three-dimensional coordinate set to visualize each part key point in the part image after automatically generating the three-dimensional coordinate set corresponding to the part key point set, so that the user can intuitively compare the position of the key point mark with the position of the part key point displayed in the part image for manual verification. Then, after determining that the position is abnormal, the position of the key point mark can be manually adjusted to calibrate the corresponding three-dimensional coordinates through the updated two-dimensional coordinates of the key point mark.

[0115] Further references Figure 3 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a three-dimensional data acquisition device. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0116] like Figure 3As shown, the three-dimensional data acquisition device 300 of some embodiments includes: an acquisition unit 301, an identification unit 302 and a generation unit 303. The acquisition unit 301 is configured to control each camera in the camera set to synchronously acquire a part image of a target part of a target object within a preset time period to obtain a part image group sequence, wherein each part image group in the part image group sequence is a part image acquired by each camera at the same time, and the camera set includes each camera set at different viewing angles of the target part; the identification unit 302 is configured to perform key point recognition on each part image in the obtained part image group sequence to obtain a two-dimensional coordinate set corresponding to a part key point set, wherein the part key point set corresponds to the target part; the generation unit 303 is configured to generate a three-dimensional coordinate set corresponding to the part key point set for each part image group in the part image group sequence according to a camera parameter information set corresponding to the camera set and each two-dimensional coordinate set corresponding to the part image group obtained.

[0117] Optionally, before the identification unit 302, the three-dimensional data acquisition device 300 may further include: a correction unit (not shown in the figure), configured to perform image correction processing on each part image in the above-mentioned part image group sequence to update the above-mentioned part image group sequence.

[0118] Optionally, the three-dimensional data acquisition device 300 may also include: a visualization unit (not shown in the figure), which is configured to perform part key point visualization processing on each part image in the above-mentioned part image group sequence that meets the verification conditions, so as to obtain each part image after visualization processing, wherein each key point mark corresponding to the above-mentioned part key point set is displayed in each part image after visualization processing.

[0119] Optionally, the three-dimensional data acquisition device 300 may further include: a determination unit, a coordinate generation unit and an update unit (not shown in the figure). The determination unit is configured to determine the two-dimensional coordinate set corresponding to the part image group that meets the above-mentioned verification condition as the target two-dimensional coordinate set in response to detecting the position adjustment operation of at least one key point mark corresponding to the key point of any part. The coordinate generation unit is configured to generate the three-dimensional coordinates corresponding to the above-mentioned key point of any part according to the target two-dimensional coordinates corresponding to the above-mentioned key point of any part in the above-mentioned target two-dimensional coordinate set and the adjusted two-dimensional coordinates of the above-mentioned at least one key point mark. The update unit is configured to update the three-dimensional coordinate set corresponding to the above-mentioned part image group according to the three-dimensional coordinates corresponding to the above-mentioned key point of any part.

[0120] Optionally, key point lines are also displayed in each part image after visualization processing.

[0121] Optionally, the three-dimensional data acquisition device 300 may further include: an image updating unit (not shown in the figure), configured to update the images of each part after visualization processing according to the three-dimensional coordinates corresponding to the key points of any of the above parts.

[0122] Optionally, the visualization unit can be further configured to: determine a three-dimensional coordinate set corresponding to the above-mentioned part image group; for each part image in the above-mentioned part image group, perform the following steps: determine each reprojected two-dimensional coordinate of the above-mentioned part image corresponding to the above-mentioned three-dimensional coordinate set, and obtain a reprojected two-dimensional coordinate set; perform part key point marking processing in the above-mentioned part image according to the above-mentioned reprojected two-dimensional coordinate set, and obtain a marked part image, wherein each two-dimensional key point marking corresponding to the above-mentioned part key point set is displayed in the marked part image; and the marked part image is displayed.

[0123] Optionally, the visualization unit may be further configured to: display a three-dimensional part model according to the determined three-dimensional coordinate set corresponding to the above-mentioned part image group, wherein the three-dimensional part model displays each three-dimensional key point mark corresponding to the above-mentioned part key point set.

[0124] Optionally, the three-dimensional data acquisition device 300 may also include: a transformation unit (not shown in the figure), which is configured to respond to detecting a transformation operation on the visualized part image or the above-mentioned three-dimensional part model, and transform the above-mentioned part image or the above-mentioned three-dimensional part model according to transformation mode information corresponding to the above-mentioned transformation operation.

[0125] Optionally, the generation unit 303 can be further configured to: determine any two cameras in the above-mentioned camera set as the initial camera set; based on the initial camera set, perform the following first iteration step: for each part key point in the above-mentioned part key point set, perform the following steps: generate initial three-dimensional coordinates corresponding to the above-mentioned part key points according to each camera parameter information corresponding to the initial camera set in the above-mentioned camera parameter information set and each two-dimensional coordinate set corresponding to the above-mentioned part image group; generate key point reprojection error information corresponding to the above-mentioned part key points according to the above-mentioned part image group and the above-mentioned initial three-dimensional coordinates; generate reprojection error information according to the obtained reprojection error information of each key point; add the generated reprojection error information to the reprojection error information set to update the reprojection error information set; in response to determining that the updated reprojection error information set satisfies the preset first iteration condition, select the reprojection error information that satisfies the preset first error condition from the updated reprojection error information set as the target reprojection error information; determine each initial three-dimensional coordinate corresponding to the above-mentioned target reprojection error information as the three-dimensional coordinate set corresponding to the above-mentioned part key point set.

[0126] Optionally, the above-mentioned iterative step also includes: in response to determining that the updated reprojection error information set does not satisfy the above-mentioned preset first iteration condition, adding an unused camera to the initial camera set to update the initial camera set, and based on the updated initial camera set, executing the above-mentioned first iteration step again.

[0127] Optionally, the generation unit 303 can be further configured to: determine each part image corresponding to the initial camera set in the above-mentioned part image group as the part image set to be measured; determine each two-dimensional coordinate corresponding to the above-mentioned part key point in each two-dimensional coordinate set corresponding to the above-mentioned part image set to be measured as the two-dimensional coordinate set to be measured; reproject the above-mentioned initial three-dimensional coordinates to each part image in the above-mentioned part image group to obtain each reprojected two-dimensional coordinate; generate key point reprojection error information corresponding to the above-mentioned part key point according to the above-mentioned each reprojected two-dimensional coordinate and the above-mentioned two-dimensional coordinate set to be measured.

[0128] Optionally, the coordinate generating unit may be further configured to: determine the various part images corresponding to the at least one key point mark in the part image group that meets the above-mentioned verification condition as a moving part image set; determine the above-mentioned moving part image set as an initial part image combination; determine the various target two-dimensional coordinates corresponding to the above-mentioned arbitrary part key points in the above-mentioned target two-dimensional coordinate set as a key point two-dimensional coordinate set; based on the initial part image combination, perform the following second iteration step: generate the initial adjusted three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points according to the various target two-dimensional coordinates in the above-mentioned key point two-dimensional coordinate set corresponding to the initial part image combination and the adjusted two-dimensional coordinates of the at least one key point mark; Based on the above-mentioned part image group and the above-mentioned initial adjusted three-dimensional coordinates, key point reprojection error information corresponding to the above-mentioned arbitrary part key points is generated; the above-mentioned key point reprojection error information is added to the key point reprojection error information set corresponding to the above-mentioned arbitrary part key points to update the key point reprojection error information set; in response to determining that the updated key point reprojection error information set satisfies the preset second error condition, the key point reprojection error information that satisfies the preset second error condition is selected from the updated key point reprojection error information set as the target key point reprojection error information; the initial adjusted three-dimensional coordinates corresponding to the above-mentioned target key point reprojection error information are determined as the three-dimensional coordinates corresponding to the above-mentioned arbitrary part key points.

[0129] Optionally, the above-mentioned second iteration step also includes: in response to determining that the updated key point reprojection error information set does not meet the above-mentioned preset second iteration condition, adding an unused part image in the part image group that meets the above-mentioned verification condition to the initial part image combination to update the initial part image combination, and based on the updated initial part image combination, executing the above-mentioned second iteration step again.

[0130] Optionally, the three-dimensional data acquisition device 300 may also include: a combination unit (not shown in the figure), which is configured to combine the reprojected two-dimensional coordinates and three-dimensional coordinates of each part key point in the part key point set corresponding to the above part image into coordinate data for each part image in the above-mentioned part image group sequence, so as to obtain a coordinate data set corresponding to the above-mentioned part image.

[0131] It is understood that the units described in the device 300 are similar to those described in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the device 300 and the units contained therein, and will not be described in detail here.

[0132] Reference below Figure 4, which shows a schematic diagram of the structure of an electronic device 400 suitable for implementing some embodiments of the present disclosure. The electronic devices in some embodiments of the present disclosure may include but are not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0133] like Figure 4 As shown, the electronic device 400 may include a processing device 401 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage device 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data required for the operation of the electronic device 400 are also stored. The processing device 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0134] Typically, the following devices may be connected to the I / O interface 405: an input device 406 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 407 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 408 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 409. The communication device 409 may allow the electronic device 400 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 4 The electronic device 400 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 4 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0135] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 409, or installed from the storage device 408, or installed from the ROM 402. When the computer program is executed by the processing device 401, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0136] It should be noted that the computer-readable medium recorded in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0137] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0138] The computer-readable medium may be included in the electronic device; or it may exist independently without being installed in the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: controls each camera in the camera set to synchronously capture part images of the target part of the target object within a preset time period to obtain a part image group sequence, wherein each part image group in the part image group sequence is a part image captured by each camera at the same time, and the camera set includes each camera set at different viewing angles of the target part; for each part image in the obtained part image group sequence, performs key point recognition on the part image to obtain a two-dimensional coordinate set corresponding to the part key point set, wherein the part key point set corresponds to the target part; for each part image group in the part image group sequence, generates a three-dimensional coordinate set corresponding to the part key point set according to the camera parameter information set corresponding to the camera set and each two-dimensional coordinate set corresponding to the part image group.

[0139] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0140] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0141] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be arranged in a processor, for example, may be described as: a processor includes an acquisition unit, an identification unit, and a generation unit. The names of these units do not constitute limitations on the units themselves in certain cases, for example, the acquisition unit may also be described as "a unit that controls each camera in a camera set to synchronously acquire a part image of a target part of a target object within a preset time period to obtain a part image group sequence".

[0142] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0143] Some embodiments of the present disclosure further provide a computer program product, including a computer program, which implements any of the above-mentioned three-dimensional data acquisition methods when executed by a processor.

[0144] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A three-dimensional data acquisition method, comprising: Within a preset time period, controlling each camera in the camera set to synchronously capture a part image of a target part of a target object to obtain a part image group sequence, wherein each part image group in the part image group sequence is a part image captured by each camera at the same time, and the camera set includes each camera set at different viewing angles of the target part; For each part image in the obtained part image group sequence, key point recognition is performed on the part image to obtain a two-dimensional coordinate set corresponding to a part key point set, wherein the part key point set corresponds to the target part; For each part image group in the part image group sequence, a three-dimensional coordinate set corresponding to the part key point set is generated according to a camera parameter information set corresponding to the camera set and each obtained two-dimensional coordinate set corresponding to the part image group.

2. The method according to claim 1, wherein: Before performing key point recognition on each part image in the obtained part image group sequence to obtain a two-dimensional coordinate set of a corresponding part key point set, the method further includes: Image correction processing is performed on each part image in the part image group sequence to update the part image group sequence.

3. The method according to claim 1, wherein: The method further comprises: For the part image groups in the part image group sequence that meet the conditions to be verified, part key point visualization processing is performed on each part image in the part image group to obtain each part image after visualization processing, wherein each key point mark corresponding to the part key point set is displayed in each part image after visualization processing.

4. The method according to claim 3, wherein: The method further comprises: In response to detecting a position adjustment operation of at least one key point mark corresponding to any part key point, determining a two-dimensional coordinate set corresponding to the part image group that meets the to-be-verified condition as a target two-dimensional coordinate set; Generate a three-dimensional coordinate corresponding to the key point of the arbitrary part according to each target two-dimensional coordinate corresponding to the key point of the arbitrary part in the target two-dimensional coordinate set and the adjusted two-dimensional coordinate of the at least one key point mark; According to the three-dimensional coordinates corresponding to the key points of the arbitrary parts, the three-dimensional coordinate set corresponding to the part image group is updated.

5. The method according to claim 4, wherein: After visualization, each part image also displays key point lines; And the method further comprises: The images of each part after visualization processing are updated according to the three-dimensional coordinates corresponding to the key points of the arbitrary part.

6. The method according to claim 3, wherein: The performing part key point visualization processing on each part image in the part image group to obtain each part image after visualization processing comprises: Determine a set of three-dimensional coordinates corresponding to the group of part images; For each part image in the part image group, the following steps are performed: Determine each reprojected two-dimensional coordinate of the part image corresponding to the three-dimensional coordinate set to obtain a reprojected two-dimensional coordinate set; According to the reprojected two-dimensional coordinate set, a part key point marking process is performed in the part image to obtain a part image after the marking process, wherein the part image after the marking process displays each two-dimensional key point mark corresponding to the part key point set; The marked part image is displayed.

7. The method according to claim 6, wherein: The performing of a part key point visualization process on each part image in the part image group to obtain each part image after visualization process further comprises: A three-dimensional part model is displayed according to the determined three-dimensional coordinate set corresponding to the part image group, wherein the three-dimensional part model displays the three-dimensional key point marks corresponding to the part key point set.

8. The method according to claim 7, wherein: The method further comprises: In response to detecting a transformation operation on the visualized part image or the three-dimensional part model, the visualized part image or the three-dimensional part model is transformed according to transformation mode information corresponding to the transformation operation.

9. The method according to claim 1, wherein: The step of generating a three-dimensional coordinate set corresponding to the part key point set according to the camera parameter information set corresponding to the camera set and each obtained two-dimensional coordinate set corresponding to the part image group comprises: Determine any two cameras in the camera set as an initial camera set; Based on the initial set of cameras, the following first iteration steps are performed: For each part key point in the part key point set, perform the following steps: Generate initial three-dimensional coordinates corresponding to the key points of the parts according to each camera parameter information corresponding to the initial camera set in the camera parameter information set and each two-dimensional coordinate set corresponding to the part image group; Generating key point reprojection error information corresponding to the key points of the part according to the part image group and the initial three-dimensional coordinates; Generate reprojection error information according to the obtained reprojection error information of each key point; Adding the generated reprojection error information to the reprojection error information set to update the reprojection error information set; In response to determining that the updated reprojection error information set satisfies a preset first iteration condition, selecting reprojection error information that satisfies the preset first error condition from the updated reprojection error information set as target reprojection error information; The initial three-dimensional coordinates corresponding to the target reprojection error information are determined as a three-dimensional coordinate set corresponding to the part key point set.

10. The method according to claim 9, wherein: The iterative step further comprises: In response to determining that the updated reprojection error information set does not satisfy the preset first iteration condition, an unused camera is added to the initial camera set to update the initial camera set, and the first iteration step is performed again based on the updated initial camera set.

11. The method according to claim 9, wherein: The step of generating key point reprojection error information corresponding to the key points of the part according to the part image group and the initial three-dimensional coordinates includes: Determine each part image corresponding to the initial camera set in the part image group as a part image set to be tested; Determine each two-dimensional coordinate corresponding to the key point of the part in each two-dimensional coordinate set corresponding to the image set of the part to be measured as the two-dimensional coordinate set to be measured; Reprojecting the initial three-dimensional coordinates to each part image in the part image group to obtain each reprojected two-dimensional coordinate; According to the respective reprojected two-dimensional coordinates and the set of two-dimensional coordinates to be measured, key point reprojection error information corresponding to the key points of the part is generated.

12. The method according to claim 4, wherein: The generating of the three-dimensional coordinates corresponding to the key point of the arbitrary part according to each target two-dimensional coordinate corresponding to the key point of the arbitrary part in the target two-dimensional coordinate set and the adjusted two-dimensional coordinates of the at least one key point mark comprises: Determine each part image corresponding to the at least one key point mark in the part image group that meets the condition to be verified as a moving part image set; Determining the moving part image set as an initial part image combination; Determine each target two-dimensional coordinate corresponding to the key point of any part in the target two-dimensional coordinate set as a key point two-dimensional coordinate set; Based on the initial part image combination, the following second iteration steps are performed: Generate initial adjusted three-dimensional coordinates corresponding to the key point of any part according to each target two-dimensional coordinate of the initial part image combination in the key point two-dimensional coordinate set and the adjusted two-dimensional coordinates of the at least one key point mark; generating key point reprojection error information corresponding to the key point of any part according to the part image group and the initial adjusted three-dimensional coordinates; Adding the key point reprojection error information to the key point reprojection error information set corresponding to the key point of the arbitrary part to update the key point reprojection error information set; In response to determining that the updated key point reprojection error information set satisfies a preset second iteration condition, selecting key point reprojection error information that satisfies the preset second error condition from the updated key point reprojection error information set as target key point reprojection error information; The initial adjusted three-dimensional coordinates corresponding to the reprojection error information of the target key point are determined as the three-dimensional coordinates corresponding to the key point of the arbitrary part.

13. The method according to claim 12, wherein: The second iteration step further comprises: In response to determining that the updated key point reprojection error information set does not satisfy the preset second iteration condition, an unused part image in the part image group that satisfies the condition to be verified is added to the initial part image combination to update the initial part image combination, and based on the updated initial part image combination, the second iteration step is performed again.

14. The method according to any one of claims 1 to 13, wherein: The method further comprises: For each part image in the part image group sequence, the reprojected two-dimensional coordinates and three-dimensional coordinates of each part key point in the part key point set corresponding to the part image are combined into coordinate data to obtain a coordinate data set corresponding to the part image.

15. A three-dimensional data acquisition device, comprising: The acquisition unit is configured to control each camera in the camera set to synchronously acquire a part image of a target part of the target object within a preset time period to obtain a part image group sequence, wherein each part image group in the part image group sequence is a part image acquired by each camera at the same time, and the camera set includes each camera set at different viewing angles of the target part; an identification unit configured to perform key point identification on each part image in the obtained part image group sequence to obtain a two-dimensional coordinate set corresponding to a part key point set, wherein the part key point set corresponds to the target part; The generating unit is configured to generate, for each part image group in the part image group sequence, a three-dimensional coordinate set corresponding to the part key point set according to a camera parameter information set corresponding to the camera set and each two-dimensional coordinate set obtained corresponding to the part image group.

16. An electronic device comprising: A camera set, the camera set comprising cameras arranged at different viewing angles of the target part, each camera being used to collect a part image of the target part; one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 14.

17. A computer readable medium having a computer program stored thereon, wherein: When the computer program is executed by a processor, the method according to any one of claims 1 to 14 is implemented.

18. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 14.

Images