Data processing method and apparatus, electronic device, and computer-readable medium

Abstract

The present disclosure provides a data processing method and apparatus, an electronic device, and a computer-readable medium. The data processing method comprises: acquiring a three-dimensional mesh to be processed and a reference three-dimensional mesh; in a preset two-dimensional planar space, rendering a first planar representation image of the three-dimensional mesh to be processed and a second planar representation image of the reference three-dimensional mesh, wherein the preset two-dimensional planar space comprises at least one anchor set; determining first position representation data and second position representation data of anchors in each anchor set, wherein the first position representation data is determined on the basis of area description information of a corresponding vertex bounding area of the anchors in the first planar representation image, and the second position representation data is determined on the basis of area description information of a corresponding vertex bounding area of the anchors in the second planar representation image; and on the basis of difference representation data between the first position representation data and the second position representation data, determining mesh performance representation data of the three-dimensional mesh to be processed. Therefore, the mesh performance evaluation effect can be improved.

Description

Data processing method, device, electronic device, and computer-readable medium

[0001] This application claims priority to Chinese Patent Application No. 202311541091.5 filed on November 17, 2023, and the contents of the above-mentioned Chinese patent application disclosure are hereby incorporated by reference in their entirety as a part of this application. Technical Field

[0002] The present disclosure relates to a data processing method, device, electronic device, and computer-readable medium. Background Art

[0003] A 3D mesh refers to a 3D structure consisting of vertices in a 3D space and edges connecting these vertices, so that the 3D mesh can be used to represent an object in a finer granularity, such as a face, an object, a building, etc.

[0004] In some application scenarios, a three-dimensional reconstruction process may be performed on an image provided by a user to obtain a three-dimensional mesh corresponding to the image, so that the three-dimensional mesh can better represent the object described by the image in three-dimensional space.

[0005] Summary of the Invention

[0006] The present disclosure provides a data processing method, apparatus, electronic device, and computer-readable medium, which can better determine the grid performance (eg, standardization, etc.) of a three-dimensional grid.

[0007] In order to achieve the above objectives, the technical solutions provided by the present disclosure are as follows:

[0008] The present disclosure provides a data processing method, the method comprising:

[0009] Obtaining a three-dimensional grid to be processed and a reference three-dimensional grid;

[0010] Drawing a first plane representation image of the 3D mesh to be processed and a second plane representation image of the reference 3D mesh in a preset 2D plane space; the first plane representation image is used to represent the state of vertices in the 3D mesh to be processed in the preset 2D plane space; the second plane representation image is used to represent the state of vertices in the reference 3D mesh in the preset 2D plane space; the preset 2D plane space includes at least one anchor point set;

[0011] Determining first position representation data and second position representation data of an anchor point in each anchor point set; the first position representation data is determined based on region description information of a region enclosed by vertices corresponding to the anchor point in the first plane representation image; and the second position representation data is determined based on region description information of a region enclosed by vertices corresponding to the anchor point in the second plane representation image;

[0012] Determine grid performance representation data of the to-be-processed three-dimensional grid based on difference representation data between the first position representation data and the second position representation data.

[0013] In one possible implementation, the process of determining the grid performance characterization data includes:

[0014] For any anchor point set, a first virtual grid and a second virtual grid are constructed using the anchor point set, and a grid performance evaluation result corresponding to the anchor point set is determined based on the grid difference representation data between the first virtual grid and the second virtual grid; the vertices in the first virtual grid are determined based on the first anchor point in the anchor point set, a vertex enclosing area corresponding to the first anchor point exists in the first plane representation image, and first description information of the vertices in the first virtual grid is determined based on the first position representation data of the first anchor point; the vertices in the second virtual grid are determined based on the second anchor point in the anchor point set, a vertex enclosing area corresponding to the second anchor point exists in the second plane representation image, and second description information of the vertices in the second virtual grid is determined based on the second position representation data of the second anchor point; the grid difference representation data is determined based on the difference representation data between the first description information of the vertices in the first virtual grid and the second description information of the vertices in the second virtual grid;

[0015] Determine mesh performance characterization data of the three-dimensional mesh to be processed based on the mesh performance evaluation result corresponding to the at least one anchor point set.

[0016] In one possible implementation, constructing the first virtual grid and the second virtual grid using the anchor point set includes:

[0017] Determine each anchor point group to be used from the anchor point set; for any anchor point group to be used, a rectangle formed by all anchor points in the anchor point group to be used satisfies a preset rectangle condition;

[0018] For any anchor point group to be used, if there is a vertex enclosing area corresponding to each anchor point in the anchor point group to be used in the first plane representation image, then the first virtual grid is constructed based on all anchor points in the anchor point group to be used and the first position representation data of all anchor points; if there are vertex enclosing areas corresponding to some anchor points in the anchor point group to be used in the first plane representation image, and the number of the some anchor points is not less than a preset number threshold, then the first virtual grid is constructed based on the some anchor points and the first position representation data of the some anchor points;

[0019] For any anchor point group to be used, if there exists a vertex enclosing area corresponding to each anchor point in the anchor point group to be used in the second plane representation image, the second virtual grid is constructed based on all the anchor points in the anchor point group to be used and the second position representation data of all the anchor points; if there exists a vertex enclosing area corresponding to some of the anchor points in the anchor point group to be used in the second plane representation image, and the number of the some anchor points is not less than a preset number threshold, the second virtual grid is constructed based on the some anchor points and the second position representation data of the some anchor points.

[0020] In one possible implementation, the anchor point group to be used includes four anchor points;

[0021] The step of constructing the first virtual grid according to all anchor points in the anchor point group to be used and the first position representation data of all anchor points includes:

[0022] constructing two triangular faces in the first virtual mesh according to the four anchor points in the anchor point group to be used and the first position representation data of the four anchor points;

[0023] The constructing the second virtual grid according to all anchor points in the to-be-used anchor point group and the second position representation data of all anchor points includes:

[0024] Two triangular faces in the second virtual mesh are constructed according to the four anchor points in the anchor point group to be used and the second position representation data of the four anchor points.

[0025] In one possible implementation, the number of the partial anchor points is three;

[0026] The constructing the first virtual grid according to the partial anchor points and the first position representation data of the partial anchor points includes:

[0027] constructing a triangular face in the first virtual mesh according to the partial anchor points and the first position representation data of the partial anchor points;

[0028] The constructing the second virtual grid according to the partial anchor points and the second position representation data of the partial anchor points includes:

[0029] A triangular face in the second virtual mesh is constructed according to the partial anchor points and the second position representation data of the partial anchor points.

[0030] In one possible implementation, the mesh difference representation data includes vertex difference representation data of a plurality of vertices; for any vertex, the vertex difference representation data of the vertex is determined based on difference representation data between the first description information of the vertex and the second description information of the vertex;

[0031] The mesh performance evaluation result is determined based on the Laplace losses corresponding to the several vertices. For any vertex, the Laplace loss corresponding to the vertex is determined based on the vertex difference characterization data of the vertex, the vertex difference characterization data of the adjacent vertices of the vertex, and the two diagonals corresponding to the edge formed by the vertex and the adjacent vertices of the vertex.

[0032] In one possible implementation, the at least one anchor point set includes N anchor point sets arranged sequentially with equal distances between rows and columns, where N is a positive integer;

[0033] The process of obtaining the nth anchor point set includes:

[0034] Determining a starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space according to preset sampling parameters and an arrangement sequence number corresponding to the nth anchor point set; the preset sampling parameters include row spacing, column spacing, and the number of anchor point sets; n is a positive integer, n≤N;

[0035] Determining at least one non-starting sampling position corresponding to the nth anchor point set according to the starting sampling position and the preset sampling parameters;

[0036] The anchor points at the starting sampling position and the anchor points at each of the non-starting sampling positions in the preset two-dimensional plane space are grouped together to obtain the nth anchor point set.

[0037] In a possible implementation manner, the vertex enclosed area is a triangular surface;

[0038] The region description information includes at least one of an identifier of the triangular face and a plane position of each vertex in the triangular face in the preset two-dimensional plane space.

[0039] In a possible implementation manner, the three-dimensional mesh to be processed is constructed using a three-dimensional reconstruction model;

[0040] After determining the grid performance characterization data of the three-dimensional grid to be processed, the method further includes:

[0041] The three-dimensional reconstructed model is updated according to the grid performance characterization data.

[0042] The present disclosure provides a data processing device, comprising:

[0043] A grid acquisition unit, used to acquire a three-dimensional grid to be processed and a reference three-dimensional grid;

[0044] An image drawing unit is configured to draw a first plane representation image of the 3D mesh to be processed and a second plane representation image of the reference 3D mesh in a preset 2D plane space; the first plane representation image is configured to represent the states of vertices in the 3D mesh to be processed in the preset 2D plane space; the second plane representation image is configured to represent the states of vertices in the reference 3D mesh in the preset 2D plane space; the preset 2D plane space includes at least one anchor point set;

[0045] a first determining unit, configured to determine first position representation data and second position representation data of an anchor point in each anchor point set; the first position representation data being determined based on region description information of a region enclosed by vertices corresponding to the anchor point in the first plane representation image; and the second position representation data being determined based on region description information of a region enclosed by vertices corresponding to the anchor point in the second plane representation image;

[0046] The second determining unit is configured to determine the grid performance representation data of the to-be-processed three-dimensional grid according to the difference representation data between the first position representation data and the second position representation data.

[0047] The present disclosure provides an electronic device, the device comprising: a processor and a memory;

[0048] The memory is used to store instructions or computer programs;

[0049] The processor is used to execute the instructions or computer programs in the memory, so that the electronic device executes the data processing method provided by the present disclosure.

[0050] The present disclosure provides a computer-readable medium, wherein the computer-readable medium stores instructions or computer programs. When the instructions or computer programs are executed on a device, the device executes the data processing method provided by the present disclosure.

[0051] The present disclosure provides a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program contains program code for executing the data processing method provided by the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure or related technologies, the following briefly introduces the drawings required for use in the embodiments or related technical descriptions. Obviously, the drawings described below are only some embodiments recorded in the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0053] FIG1 is a schematic diagram of a three-dimensional grid provided by an embodiment of the present disclosure;

[0054] FIG2 is a schematic diagram of another three-dimensional grid provided by an embodiment of the present disclosure;

[0055] FIG3 is a flow chart of a data processing method provided by an embodiment of the present disclosure;

[0056] FIG4 is a schematic diagram of a plane representation image provided by an embodiment of the present disclosure;

[0057] FIG5 is a schematic diagram of another planar representation image provided by an embodiment of the present disclosure;

[0058] FIG6 is a schematic diagram of constructing a triangular face provided by an embodiment of the present disclosure;

[0059] FIG7 is a schematic diagram of a virtual grid provided by an embodiment of the present disclosure;

[0060] FIG8 is a schematic diagram of two vertices having a connection relationship provided by an embodiment of the present disclosure;

[0061] FIG9 is a schematic structural diagram of a data processing device provided by an embodiment of the present disclosure; and

[0062] FIG10 is a schematic structural diagram of an electronic device provided in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0063] After research, it was found that in some application scenarios, for the reconstructed three-dimensional grid (for example, the three-dimensional grid shown in Figure 1), it may be necessary to evaluate the accuracy and / or standardization of the three-dimensional grid. Among them, the accuracy is used to describe whether the reconstructed three-dimensional grid can accurately fit the actual object (for example, face, etc.) described by the image data provided by the user. The standardization is used to describe whether the reconstructed three-dimensional grid is normalized; and the present disclosure does not limit the implementation method of the normalization. For example, in some application scenarios, the normalization can be subjectively judged with the help of checkerboard calibration plate material to determine whether the two-dimensional projection of each black and white grid and its boundary line is horizontal and vertical after the checkerboard calibration plate material is rendered onto the three-dimensional grid. For example, for the three-dimensional grid shown in Figure 2, after the checkerboard calibration plate material is rendered onto the three-dimensional grid, the two-dimensional projection of each black and white grid and its boundary line is horizontal and vertical, so that the three-dimensional grid is completely normalized. It can be seen that the standardization evaluation process based on the checkerboard calibration plate material can only indicate whether a three-dimensional grid meets the specifications to a certain extent, and cannot accurately characterize the degree of standardization of a three-dimensional grid, so that the evaluation process cannot quantify the standardization and thus leads to poor evaluation results.

[0064] Based on the content of the above paragraph, it can be seen that in order to better improve the performance evaluation effect of three-dimensional grids, the present disclosure provides a data processing method, which includes: after obtaining the three-dimensional grid to be processed (for example, a reconstructed three-dimensional grid) and the reference three-dimensional grid (for example, a pre-set normalized three-dimensional grid template), drawing a first plane representation image of the three-dimensional grid to be processed and a second plane representation image of the reference three-dimensional grid in a preset two-dimensional plane space (for example, a plane blank image), so that the first plane representation image is used to represent the state of the vertices in the three-dimensional grid to be processed in the preset two-dimensional plane space (for example, the position distribution state of the vertices, and the connection state between different vertices, etc.), and the second plane representation image is used to represent the state of the vertices in the reference three-dimensional grid in the preset two-dimensional plane space, so that when When the preset two-dimensional plane space includes at least one anchor point set, the first position representation data and the second position representation data of each anchor point in the anchor point set are determined, wherein the first position representation data is determined based on the area description information of the vertex enclosed area corresponding to the anchor point in the first plane representation image, so that the first position representation data can, to a certain extent, represent the state of the three-dimensional mesh to be processed on the anchor point; the second position representation data is determined based on the area description information of the vertex enclosed area corresponding to the anchor point in the second plane representation image, so that the second position representation data can, to a certain extent, represent the state of the reference three-dimensional mesh on the anchor point; then, based on the difference representation data between the first position representation data and the second position representation data, the mesh performance representation data of the three-dimensional mesh to be processed is determined. Among them, because the difference characterization data can represent the difference between the three-dimensional grid to be processed and the reference three-dimensional grid, the grid performance characterization data determined based on the difference characterization data can represent the grid performance presented by the three-dimensional grid to be processed relative to the reference three-dimensional grid, so that the grid performance characterization data can represent the grid performance quantification results presented by the three-dimensional grid to be processed under one or more grid performance evaluation dimensions (for example, dimensions such as standardization). In this way, the purpose of quantitative evaluation processing of a three-dimensional grid under these grid performance evaluation dimensions can be achieved, thereby effectively avoiding defects (for example, poor grid performance evaluation effect, etc.) caused by the inability to quantify a certain grid performance evaluation dimension (for example, dimensions such as standardization) or the poor quantification scheme for a certain grid performance evaluation dimension (for example, dimensions such as accuracy), which is conducive to improving the grid performance evaluation effect.

[0065] In addition, the present disclosure does not limit the execution subject of the data processing method provided in the embodiments of the present disclosure. For example, the data processing method provided in the embodiments of the present disclosure can be applied to a terminal device or a server. For another example, the data processing method provided in the embodiments of the present disclosure can also be implemented by means of a data interaction process between a terminal device and a server. The terminal device can be a smart phone, a computer, a personal digital assistant (PDA), a tablet computer, etc. The server can be a standalone server, a cluster server, or a cloud server.

[0066] In order to enable those skilled in the art to better understand the solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the embodiments described are only part of the embodiments of the present disclosure, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present disclosure without making any creative efforts shall fall within the scope of protection of the present disclosure.

[0067] To better understand the technical solutions provided by the present disclosure, the data processing method provided by the present disclosure is described below with reference to some figures. As shown in Figure 3, the data processing method provided by the embodiment of the present disclosure includes the following steps S1-S4. Figure 3 is a flow chart of a data processing method provided by the embodiment of the present disclosure.

[0068] S1: Obtain a 3D mesh to be processed and a reference 3D mesh.

[0069] The "processed 3D mesh" refers to the 3D mesh for which mesh performance evaluation is required. The 3D mesh is used to describe the state of the processed object (e.g., a face, an object, a living being, etc.) in 3D space. The "processed object" refers to the object described by the 3D mesh. This disclosure does not limit the scope of the processed object; for example, the processed object can be a face, an object, a building, or a living being.

[0070] It should be noted that the present disclosure does not limit the implementation method of the three-dimensional grid. For example, it can be implemented using a mesh grid (for example, the mesh grid shown in Figure 1) so that the three-dimensional grid includes multiple vertices, state representation data of each vertex, and edges for connecting different vertices. Among them, for any vertex in the three-dimensional grid, the state representation data of the vertex can be used to describe the state of the vertex, and the present disclosure does not limit the implementation method of the state representation data of the vertex. For example, it can at least include two-dimensional plane position description data (for example, UV coordinates, etc.) and three-dimensional space position description data (for example, XYZ coordinates, etc.).

[0071] In addition, the present disclosure does not limit the implementation method of the above-mentioned three-dimensional grid to be processed. For ease of understanding, two examples are used for illustration below.

[0072] Example 1: In some application scenarios (for example, scenarios such as performing normative quantitative processing on a reconstructed three-dimensional mesh), the three-dimensional mesh to be processed mentioned above may refer to a three-dimensional mesh obtained by performing three-dimensional reconstruction processing on a certain image (for example, an image input by a user), so that the mesh performance evaluation task of the reconstructed three-dimensional mesh can be completed subsequently through mesh performance evaluation processing on the three-dimensional mesh to be processed.

[0073] It can be seen that, under a possible implementation method, the acquisition process of the above-mentioned three-dimensional mesh to be processed can be: after obtaining the object description image provided by the user (for example, a facial description image, etc.), the object description image can be subjected to three-dimensional reconstruction processing to obtain the three-dimensional mesh to be processed, so that the three-dimensional mesh to be processed can represent the state of the object described by the object description image (that is, the above-mentioned object to be processed) in the three-dimensional space. Among them, the object description image is used to describe the object to be processed; and because the object description image is a two-dimensional image, the object description image can be used to describe the state of the object to be processed in a two-dimensional plane. In addition, the present disclosure does not limit the provision method of the object description image. For example, the object description image may refer to an image specified by an image selection operation triggered by the user. For another example, the object description image may refer to an image captured by the user with the help of a certain image acquisition device. For another example, the object description image may refer to an image input by the user with the help of a certain input device.

[0074] Example 2: In some application scenarios (e.g., training a 3D reconstruction model), the 3D mesh to be processed may refer to a 3D mesh obtained by performing 3D reconstruction processing on an image using the 3D reconstruction model, so that the model performance of the 3D reconstruction model can be determined by subsequently performing mesh performance evaluation on the 3D mesh to be processed. The 3D reconstruction model is used to perform 3D reconstruction processing on the input data of the 3D reconstruction model (e.g., an image), and the embodiments of this disclosure do not limit the implementation of the 3D reconstruction model.

[0075] It can be seen that under one possible implementation method, the acquisition process of the above-mentioned three-dimensional mesh to be processed can be: for any round of training process, after obtaining the sample image, the three-dimensional reconstruction model can be used to perform three-dimensional reconstruction processing on the sample image to obtain and output the three-dimensional mesh to be processed, so that the three-dimensional mesh to be processed can represent the three-dimensional reconstruction result for the sample image. Among them, the sample image refers to the image required for training the three-dimensional reconstruction model; and the present disclosure does not limit the implementation method of the sample image. For example, the sample image can refer to any image in the sample image set required for training the three-dimensional reconstruction model. In addition, the present disclosure does not limit the acquisition process of the sample image.

[0076] The reference 3D mesh is the 3D mesh used as a reference when evaluating the mesh performance of the 3D mesh to be processed. This allows the mesh performance of the 3D mesh to be measured using the reference 3D mesh as a benchmark by calculating the relative difference between the 3D mesh to be processed and the reference 3D mesh. This reference 3D mesh can be used to describe the state of the object in 3D space when it is in a standard state across one or more mesh performance evaluation dimensions.

[0077] In addition, the present disclosure does not limit the implementation method of the reference 3D grid described above. For example, in some application scenarios (e.g., scenarios for normative evaluation and processing of 3D grids), the reference 3D grid can refer to a pre-set normalized 3D grid template (e.g., the 3D grid shown in FIG2 ), so that the reference 3D grid can represent the state of an object in 3D space when the object is normalized. The normalized 3D grid template is used to represent the characteristics of a 3D grid in a normalized state. Furthermore, the present disclosure does not limit the method for obtaining the normalized 3D grid template; for example, it can be pre-set by relevant personnel.

[0078] For example, in some application scenarios (e.g., for 3D mesh accuracy assessment), the reference 3D mesh can be a pre-set label 3D mesh for the previously processed 3D mesh, so that the label 3D mesh accurately represents the actual state of the previously processed object in 3D space. The label 3D mesh accurately represents the actual state of the previously processed object in 3D space. Furthermore, this disclosure does not limit the method for obtaining the label 3D mesh; for example, it can be pre-set by relevant personnel.

[0079] In addition, the present disclosure does not limit the method for obtaining the above-mentioned reference three-dimensional grid. For example, the reference three-dimensional grid can be pre-set by relevant personnel based on the grid performance evaluation requirements of actual application scenarios, which is conducive to improving the grid performance evaluation effect.

[0080] Based on the relevant content of S1 above, it can be seen that for the three-dimensional grid to be processed, if you want to perform a grid performance evaluation on the three-dimensional grid to be processed, you not only need to obtain the three-dimensional grid to be processed, but also need a reference three-dimensional grid corresponding to the three-dimensional grid to be processed, so that you can subsequently use the reference three-dimensional grid as a benchmark to measure the grid performance of the three-dimensional grid to be processed.

[0081] S2: Draw a first plane representation image of the three-dimensional mesh to be processed and a second plane representation image of the reference three-dimensional mesh in a preset two-dimensional plane space; the first plane representation image is used to represent the state of the vertices in the three-dimensional mesh to be processed in the preset two-dimensional plane space; the second plane representation image is used to represent the state of the vertices in the reference three-dimensional mesh in the preset two-dimensional plane space; the preset two-dimensional plane space includes at least one anchor point set.

[0082] The preset two-dimensional plane space refers to the two-dimensional plane space required to be used when performing grid performance evaluation processing on a three-dimensional grid.

[0083] In addition, the present disclosure does not limit the implementation of the above preset two-dimensional plane space. For example, in some application scenarios, the preset two-dimensional plane space can adopt a preset r h ×r w The implementation is performed on a planar blank image I. h Indicates the size of the preset two-dimensional plane space in one dimension (for example, pixel row direction or height); r w Indicates the size of the preset two-dimensional plane space in another dimension (for example, pixel column direction or width); and the present disclosure does not limit the r h and r w For example, in some application scenarios, in order to take into account the application scope and grid performance evaluation effect, r h =r w =1280. That is, in a possible implementation, the preset two-dimensional plane space can be implemented using a 1280×1280 plane blank image I.

[0084] In addition, for the above presupposed two-dimensional plane space (for example, r h ×r w For example, a large number of anchor points are distributed in the preset two-dimensional plane space, so that different anchor points are used to represent different positions in the preset two-dimensional plane space. h ×r wWhen the planar blank image I is implemented, the anchor points distributed in the preset two-dimensional plane space may refer to the pixel points appearing in the planar blank image I, so that the preset two-dimensional plane space is distributed with r h ×r w Anchor points.

[0085] In addition, in order to better improve the effect of grid performance evaluation, the anchor points in the above preset two-dimensional plane space can be used to construct at least one anchor point set, so that each anchor point set can represent some anchor points in the preset two-dimensional plane space, and there are differences between the anchor points represented by different anchor point sets (for example, there is no intersection between different anchor point sets, etc.), so that different anchor point sets can describe the preset two-dimensional plane space from different angles, so that these anchor point sets can be used to more accurately and comprehensively determine the differences between the above-mentioned three-dimensional grid to be processed and the reference three-dimensional grid in the preset two-dimensional plane space.

[0086] Furthermore, the present disclosure does not limit the method for constructing the at least one anchor point set. For example, in some application scenarios, the construction process of the at least one anchor point set may specifically include: performing random partitioning processing on all anchor points in the preset two-dimensional plane space to obtain at least one anchor point set, such that the number of anchor points in different anchor point sets is the same, and there is no intersection between different anchor point sets. For another example, in some application scenarios, the construction process of the at least one anchor point set may specifically include: performing full-row and full-column partitioning processing on all anchor points in the preset two-dimensional plane space to obtain at least one anchor point set, such that the number of anchor point rows and the number of anchor point columns in different anchor point sets are the same, and there is no intersection between different anchor point sets.

[0087] In fact, in order to better improve the effect of grid performance evaluation, at least one anchor point set with equal spacing between rows and columns can be constructed, so that each anchor point set can represent the global state of the preset two-dimensional plane space mentioned above, and different anchor point sets are used for the global state of the preset two-dimensional plane space at different angles. Based on this, it can be seen that in one possible implementation, when the preset two-dimensional plane space includes at least one anchor point set, the at least one anchor point set can include N anchor point sets arranged in sequence with equal spacing between rows and columns, so that the number of anchor points in each anchor point set is equal, so that the number of anchor points in each anchor point set can be determined according to the following formula (1), where N is a positive integer.

[0088] Where, E number Indicates the number of anchor points in any anchor point set; r h Indicates the size of the above-preset two-dimensional plane space in one dimension (for example, pixel row direction or height); r wIndicates the size of the preset two-dimensional plane space in another dimension (for example, pixel column direction or width); s h Indicates line spacing; s w Indicates the spacing between columns.

[0089] It should be noted that equal spacing between rows means that for any anchor point set, the distances between each pair of closest anchor point rows in that anchor point set are equal, and the distance between any pair of closest anchor point rows in that anchor point set is equal to the distance between any pair of closest anchor point rows in other anchor point sets. Similarly, equal spacing between columns means that for any anchor point set, the distances between each pair of closest anchor point columns in that anchor point set are equal, and the distance between any pair of closest anchor point columns in that anchor point set is equal to the distance between any pair of closest anchor point columns in other anchor point sets.

[0090] Based on the above formula (1), it can be seen that in a possible implementation, when the preset two-dimensional plane space includes N anchor point sets arranged in sequence with equal spacing between rows and columns, the number of anchor points in the nth anchor point set is based on the preset sampling parameters (for example, the above s h and above w etc.) and the size parameters of the preset two-dimensional plane space (e.g., r h and r w etc.); n is a positive integer, n≤N. Among them, the preset sampling parameters refer to the parameters (such as sampling spacing, number of anchor point sets, etc.) required to be used when performing equidistant sampling processing on the preset two-dimensional plane space above; the present disclosure does not limit the preset sampling parameters. For example, when performing equidistant sampling processing between rows and equidistant sampling between columns on the preset two-dimensional plane space, the preset sampling parameters may include at least one or more of row spacing, column spacing and number of anchor point sets. The row spacing refers to the parameter required to be used when performing equidistant sampling processing in the direction of the anchor point row on the preset two-dimensional plane space, so that the row spacing can represent the distance between the two anchor point rows that are closest to each other in any anchor point set, so that the row spacing can represent the distance between the two anchor points that are closest to each other in the row direction in any anchor point set. The column spacing refers to the parameter required for performing equally spaced sampling in the anchor point column direction within the preset two-dimensional plane space, such that the column spacing can represent the distance between the two closest anchor point columns within any anchor point set, and thus the column spacing can represent the distance between the two closest anchor points within any anchor point set in the column direction. The number of anchor point sets refers to the number of anchor point sets sampled from the preset two-dimensional plane space, that is, N.

[0091] It should be noted that, for the number of anchor point sets in the upper section, the larger the number of anchor point sets, the more accurate the mesh performance measurement for the three-dimensional mesh, but the greater the computational effort. However, for the row and column spacing in the upper section, the larger these two values, the greater the distance between any two closest anchor points in an anchor point set, resulting in a coarser mesh performance measurement. Therefore, to better meet the mesh performance evaluation requirements of different application scenarios, the preset sampling parameters described above can be set based on the mesh performance evaluation requirements of the actual application scenario.

[0092] It should also be noted that the present disclosure does not limit the method for obtaining the above-mentioned preset sampling parameters. For example, they can be set in advance according to the application scenario. In particular, the preset sampling parameters can be set in advance according to the actual application scenario so that the preset sampling parameters meet the grid performance measurement requirements under the application scenario.

[0093] Based on the above content, it can be seen that in some application scenarios, at least one of the above anchor point sets can be obtained by performing equal-interval sampling processing on all anchor points in the above preset two-dimensional plane space (for example, equal spacing between rows and equal spacing between columns, etc.), so that each anchor point set can represent the global state of the preset two-dimensional plane space to a certain extent.

[0094] In addition, the present disclosure does not limit the acquisition process of the above-mentioned at least one anchor point set. For example, in some application scenarios, when the above-preset two-dimensional plane space includes at least one anchor point set, and the at least one anchor point set may include N anchor point sets arranged in sequence with equal distances between rows and columns, the acquisition process of the nth anchor point set may specifically include the following steps 11 to 13.

[0095] Step 11: Determine the starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space according to the preset sampling parameters and the arrangement sequence number corresponding to the nth anchor point set; the preset sampling parameters include row spacing, column spacing and the number of anchor point sets; n is a positive integer, n≤N.

[0096] For details on the preset sampling parameters, please refer to the above text, which will not be repeated here for the sake of brevity.

[0097] The nth anchor point set refers to the anchor point set in the nth arrangement position in at least one of the above anchor point sets, where n is a positive integer, n≤N.

[0098] In addition, for the nth anchor point set above, the arrangement number corresponding to the nth anchor point set is used to indicate the arrangement position of the nth anchor point set in at least one of the above anchor point sets; and the present disclosure does not limit the implementation method of the arrangement number corresponding to the nth anchor point set. For example, in some application scenarios, the arrangement number corresponding to the nth anchor point set can be implemented using the value n.

[0099] Furthermore, for the nth anchor point set described above, the starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space refers to the first sampling position required to be used when sampling the nth anchor point set from the preset two-dimensional plane space, such that the starting sampling position can represent the position of the first anchor point in the nth anchor point set in the preset two-dimensional plane space. The first anchor point refers to the first anchor point in the nth anchor point set.

[0100] In addition, the present disclosure does not limit the method for obtaining the "starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space" in the previous paragraph. For example, in some application scenarios, the "starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space" can be determined according to pre-set rules.

[0101] For example, in some application scenarios, the representation effect (e.g., comprehensiveness) of the global state of the preset two-dimensional plane space by at least one anchor point set can be improved by ensuring that all anchor point sets can cover all anchor points in the preset two-dimensional plane space as comprehensively as possible. Based on this, it can be seen that the "starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space" can be determined according to the following formula (2).

[0102] In the formula, (δ y ,δ x ) represents the above “the starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space”; δ y represents the position coordinates of the “starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space” in the first dimension (e.g., row direction); δ x represents the position coordinates of the "starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space" in the second dimension (for example, the column direction); n represents the arrangement number corresponding to the nth anchor point set; N represents the number of the above anchor point sets; s h Indicates line spacing; s w Indicates column spacing; mod indicates remainder processing.

[0103] Based on the relevant content of step 11 above, it can be known that in some application scenarios, when the above preset two-dimensional plane space adopts r h ×r w When implementing the planar blank image I, starting from the pixel position (0,0) of the planar blank image I, N anchor point sets with equal spacing between rows and columns are determined, so that the number of anchor points in each anchor point set is E shown in the above formula (1). number, and the starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space is (δ y ,δ x ) so that the positions of the other anchor points in the nth anchor point set except the first anchor point can be deduced based on the starting sampling position, where n is a positive integer, n≤N, so that the N anchor point sets finally obtained can fully cover all anchor points in the preset two-dimensional plane space.

[0104] Step 12: Determine at least one non-starting sampling position corresponding to the nth anchor point set based on the starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space and the above preset sampling parameters; n is a positive integer, n≤N.

[0105] The non-starting sampling position is used to describe the positions of all anchor points in the nth anchor point set, excluding the anchor point at the starting sampling position. Therefore, in one possible implementation, the jth non-starting sampling position is used to describe the position of the jth non-first anchor point in the nth anchor point set (i.e., the jth anchor point among all anchor points in the anchor point set, excluding the first anchor point) in the preset two-dimensional plane space. j is a positive integer, and j≤N-1.

[0106] In addition, the present disclosure does not limit the implementation of the above step 12. For example, the specific implementation of step 12 may be: for the j-th anchor point in the above n-th anchor point set except the anchor point at the above starting sampling position, first, according to the size parameters of the preset two-dimensional plane space (for example, the above r h and r w etc.), the above preset sampling parameters (for example, the above N, s h 、s w etc.), and the corresponding arrangement number of the j-th anchor point in the n-th anchor point set (for example, the value of j), to obtain the position derivation parameters (for example, j, s shown in the following formulas (3)-(5) h 、c h 、c w 、s w , etc.), so that the position derivation parameter can represent the parameters required for deducing the position of the j-th anchor point based on the above “the starting sampling position corresponding to the n-th anchor point set in the preset two-dimensional plane space”; and then based on the starting sampling position (for example, the above (δ y ,δ x)), the position derivation parameter, and a preset position derivation formula (e.g., formula (3) below), determine the jth non-starting sampling position corresponding to the nth anchor point set, so that the jth non-starting sampling position can represent the position of the jth anchor point in the preset two-dimensional plane space. j is a positive integer, j≤N-1.

[0107] Where, Indicates the jth non-starting sampling position corresponding to the nth anchor point set above, so that It is used to indicate the position of the j-th anchor point in the n-th anchor point set except the anchor point at the starting sampling position above in the preset two-dimensional plane space; Indicates the position coordinates of the j-th non-starting sampling position corresponding to the n-th anchor point set in the first dimension (for example, row direction); represents the position coordinates of the jth non-starting sampling position corresponding to the nth anchor point set in the second dimension (e.g., column direction); j is determined based on the permutation sequence number corresponding to the jth anchor point in the nth anchor point set, so that j can represent the permutation position of the jth anchor point among all non-first anchor points in the nth anchor point set (i.e., all anchor points in the nth anchor point set except the first anchor point at the starting sampling position above); c h Indicates the number of anchor rows in an anchor set; c w Indicates the number of anchor columns in an anchor set; s h Indicates line spacing; s w Indicates column spacing; δ y represents the position coordinates of the “starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space” in the first dimension (e.g., row direction); δ x represents the position coordinates of the “starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space” in the second dimension (e.g., row direction); r h Indicates the size of the above-preset two-dimensional plane space in one dimension (for example, pixel row direction or height); r w It represents the size of the preset two-dimensional plane space in another dimension (for example, pixel column direction or width).

[0108] It should be noted that the present disclosure does not limit the method for determining j in the above paragraph. For example, in some application scenarios, when the permutation number corresponding to the first anchor point in the nth anchor point set is 0, j can be directly implemented using the permutation number corresponding to the jth non-first anchor point in the nth anchor point set. For another example, in some application scenarios, when the permutation number corresponding to the first anchor point in the nth anchor point set is 1, j = the permutation number corresponding to the jth anchor point in the nth anchor point set - 1, so that j can represent the permutation position of the jth non-first anchor point among all non-first anchor points in the nth anchor point set.

[0109] Based on the relevant content of step 12 above, it can be known that for the nth anchor point set, after obtaining the starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space, some non-starting sampling positions corresponding to the nth anchor point set in the preset two-dimensional plane space can be derived based on the starting sampling position, the size parameters of the preset two-dimensional plane space, and the above preset sampling parameters, so that each non-starting sampling position can represent the position of the corresponding non-first anchor point in the nth anchor point set in the preset two-dimensional plane space.

[0110] Step 13: After obtaining the starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space and at least one non-starting sampling position corresponding to the nth anchor point set, the anchor points at the starting sampling position and the anchor points at each non-starting sampling position in the preset two-dimensional plane space are grouped together to obtain the nth anchor point set; n is a positive integer, n≤N.

[0111] In the present disclosure, for the nth anchor point set, after obtaining the starting sampling position and at least one non-starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space, the anchor points at the starting sampling position and the anchor points at each non-starting sampling position in the preset two-dimensional plane space can be collected to obtain the nth anchor point set, so that the position of the first anchor point in the nth anchor point set is the starting sampling position, and the positions of the other anchor points in the nth anchor point set except the first anchor point are the non-starting sampling positions (for example, the position of the jth non-first anchor point in the nth anchor point set is the jth non-starting sampling position above, j is a positive integer, j≤N-1).

[0112] Based on the relevant contents of steps 11 to 13 above, it can be known that in some application scenarios, for the preset two-dimensional plane space above, the starting sampling position and some non-starting sampling positions involved in each anchor point set can be determined based on the preset sampling parameters; then, based on the starting sampling position and some non-starting sampling positions involved in each anchor point set, all anchor points in each anchor point set can be determined. In this way, equidistant sampling in both the row and column directions can be achieved for the preset two-dimensional plane space, so as to obtain multiple anchor point sets with equidistant rows and equidistant columns. Among them, because the anchor points in each anchor point set are distributed within the global range of the preset two-dimensional plane space, each anchor point set can respectively represent the global sampling results for the preset two-dimensional plane space, so that each anchor point set can represent the global state of the preset two-dimensional plane space to a certain extent; and because different anchor point sets can describe the preset two-dimensional plane space from different angles, these anchor point sets can describe the global state of the preset two-dimensional plane space from multiple angles, so that when performing grid performance evaluation processing on a certain three-dimensional grid based on these anchor point sets, the performance achieved by the three-dimensional grid can be analyzed from as many angles as possible, which is conducive to improving the grid performance evaluation effect.

[0113] The first plane representation image is used to represent the state of the vertices in the three-dimensional mesh to be processed within a preset two-dimensional plane space (e.g., the vertex position distribution state, the connection state between different vertices, etc.). For example, the first plane representation image can be implemented using the image shown in Figure 4 or the image shown in Figure 5.

[0114] In addition, regarding the first planar representation image of the three-dimensional mesh to be processed described above, the first planar representation image can be obtained by plotting the two-dimensional planar position description data (e.g., UV coordinates, etc.) of each vertex in the three-dimensional mesh to be processed and all edges used to connect the vertices into the preset two-dimensional plane space, so that the first planar representation image can represent the position of each vertex in the three-dimensional mesh to be processed in the preset two-dimensional plane space and the connection relationship between different vertices, thereby enabling the first planar representation image to represent the state (e.g., position, etc.) of the closed area (e.g., a triangle) enclosed by some vertices in the three-dimensional mesh to be processed in the preset two-dimensional plane space. It can be seen that the first planar representation image can include multiple vertices (e.g., vertex 1 shown in FIG5 ), edges used to connect different vertices (e.g., the edge used to connect vertex 1 and vertex 2 shown in FIG5 ), and some vertex enclosed areas (e.g., the triangle enclosed by vertex 1, vertex 2, and vertex 3 shown in FIG5 ), so that the first planar representation image can represent the characteristics of the three-dimensional mesh to be processed.

[0115] It should be noted that the present disclosure does not limit the implementation method of "drawing" in the above paragraph. For example, it can be implemented by any existing or future method that can draw a three-dimensional grid onto a two-dimensional plane (for example, a flat blank image, etc.).

[0116] The second plane representation image is used to represent the states of the vertices in the above-mentioned referenced three-dimensional grid in a preset two-dimensional plane space (eg, the distribution state of the vertex positions, the connection state between different vertices, etc.).

[0117] In addition, for the second plane representation image of the reference three-dimensional mesh mentioned above, the second plane representation image can be obtained by drawing the two-dimensional plane position description data (e.g., UV coordinates, etc.) of each vertex in the reference three-dimensional mesh and all edges used to connect the vertices into the preset two-dimensional plane space, so that the second plane representation image can represent the position of each vertex in the reference three-dimensional mesh in the preset two-dimensional plane space and the connection relationship between different vertices, so that the second plane representation image can represent the state (e.g., position, etc.) of the closed area (e.g., a triangle face) surrounded by some vertices in the reference three-dimensional mesh in the preset two-dimensional plane space. It can be seen that for the second plane representation image, the second plane representation image can include multiple vertices, edges used to connect different vertices, and some vertex-enclosed areas, so that the second plane representation image can represent the characteristics of the reference three-dimensional mesh itself. It should be noted that the method for obtaining the second plane representation image mentioned above is similar to the method for obtaining the first plane representation image mentioned above. For the sake of brevity, it will not be repeated here.

[0118] Based on the relevant content of S2 above, it can be known that after obtaining the three-dimensional grid to be processed and the reference three-dimensional grid, a first plane representation image of the three-dimensional grid to be processed and a second plane representation image of the reference three-dimensional grid can be drawn respectively in a preset two-dimensional plane space (for example, the plane blank image I above), so that the first plane representation image can represent the state of the three-dimensional grid to be processed in the preset two-dimensional plane space, and the second plane representation image can represent the state of the reference three-dimensional grid in the preset two-dimensional plane space, so that the first plane representation image can represent the characteristics of the three-dimensional grid to be processed itself, and the second plane representation image can represent the characteristics of the reference three-dimensional grid itself, so that the difference between the three-dimensional grid to be processed and the reference three-dimensional grid can be determined subsequently with the help of at least one anchor point set of the preset two-dimensional plane space, the first plane representation image and the second plane representation image, so as to realize the grid performance evaluation processing of a three-dimensional grid using at least one anchor point set of the preset two-dimensional plane space. Among them, because different anchor point sets can describe the preset two-dimensional plane space at different angles, when performing grid performance evaluation on a certain three-dimensional grid based on these anchor point sets, the performance achieved by the three-dimensional grid can be analyzed from as many angles as possible, which is conducive to improving the grid performance evaluation effect.

[0119] S3: Determine the first position representation data and the second position representation data of the anchor point in each anchor point set; the first position representation data is determined based on the area description information of the vertex enclosed area corresponding to the anchor point in the first plane representation image; the second position representation data is determined based on the area description information of the vertex enclosed area corresponding to the anchor point in the second plane representation image.

[0120] In the present disclosure, for the i-th anchor point in the n-th anchor point set (for example, anchor point 1 shown in Figure 5), if the position of the i-th anchor point in the preset space belongs to the area occupied by a vertex enclosed area in the first plane representation image above (for example, the triangle surface enclosed by vertex 1, vertex 2 and vertex 3 shown in Figure 5) in the preset space, then it can be determined that the i-th anchor point falls within this vertex enclosed area, so that a correspondence between the i-th anchor point and this vertex enclosed area can be constructed, so that the correspondence can indicate that this vertex enclosed area is the vertex enclosed area corresponding to the i-th anchor point in the first plane representation image, so that the first position representation data of the i-th anchor point can be determined based on the area description information of the vertex enclosed area corresponding to the i-th anchor point in the first plane representation image, so that the first position representation data can represent the characteristics of the three-dimensional mesh to be processed under the i-th anchor point. The region description information is used to describe the vertex enclosing region; and the present disclosure does not limit the region description information. For example, it may include the region identifier of the vertex enclosing region and at least one of the plane positions of each vertex in the vertex enclosing region in a preset two-dimensional plane space. The region identifier is used to uniquely identify the vertex enclosing region; and the present disclosure does not limit the implementation method of the region identifier. For example, the region identifier may refer to the region number corresponding to the vertex enclosing region in the three-dimensional mesh to be processed. Wherein, i is a positive integer, i≤I, I is a positive integer, I represents the number of anchor points in the nth anchor point set, n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0121] It should be noted that the present disclosure does not limit the implementation method of the step of "determining the first position representation data of the i-th anchor point based on the area description information of the vertex enclosing area corresponding to the i-th anchor point in the first plane representation image" in the previous paragraph. For example, when the area description information includes the area identifier of the vertex enclosing area and the planar positions of each vertex in the vertex enclosing area in the preset two-dimensional plane space, the center of gravity coordinates of the i-th anchor point in the vertex enclosing area can be calculated based on the planar positions of each vertex in the vertex enclosing area in the preset two-dimensional plane space (for example, the center of gravity coordinates shown in formulas (6)-(7) below), so that the center of gravity coordinates can indicate which weight parameters need to be used when linearly representing the position of the i-th anchor point using the positions of all vertices in the vertex enclosing area; and then, based on the area identifier of the vertex enclosing area and the center of gravity coordinates of the i-th anchor point in the vertex enclosing area, the first position representation data of the i-th anchor point is determined, so that the first position representation data includes the area identifier and the center of gravity coordinates.

[0122] Where, Represents the centroid coordinates of the i-th anchor point in the n-th anchor point set within the area enclosed by its corresponding vertex; Indicates the position of the i-th anchor point in the preset two-dimensional plane space; represents the position of the t-th vertex in the vertex enclosed area in the preset two-dimensional plane space, t is a positive integer, t≤T, T is a positive integer, and T represents the number of vertices in the vertex enclosed area; Indicates the weighted weight corresponding to the area enclosed by the vertex (for example, linear weight, etc.).

[0123] Based on the above content, it can be seen that in one possible implementation, when the areas enclosed by each vertex in the first plane representation image above are all triangular faces, for the i-th anchor point in the n-th anchor point set (for example, anchor point 1 shown in Figure 5), if the i-th anchor point falls within a triangular face in the first plane representation image (for example, the triangular face enclosed by vertex 1, vertex 2 and vertex 3 shown in Figure 5), a correspondence between the anchor point and the triangular face is constructed, so that the first position representation data of the i-th anchor point (for example, the identity of the triangular face and at least one of the planar positions of each vertex in the triangular face in a preset two-dimensional plane space) can be determined based on the area description information of the triangular face corresponding to the i-th anchor point in the first plane representation image (for example, the identity of the triangular face and the centroid coordinates of the i-th anchor point in the triangular face, etc.), so that the first position representation data can represent the characteristics of the three-dimensional mesh to be processed under the i-th anchor point. It should be noted that the present disclosure does not limit the implementation method of the identification of the triangular face. For example, it can specifically refer to the number of the area corresponding to the triangular face in the three-dimensional mesh to be processed. Wherein, i is a positive integer, i≤I, I is a positive integer, I represents the number of anchor points in the nth anchor point set, and n is a positive integer, n≤N, N represents the number of anchor point sets.

[0124] Similarly, for the i-th anchor point in the n-th anchor point set, if the position of the i-th anchor point in the preset space falls within the area occupied by a vertex enclosing region in the second planar representation image, then the i-th anchor point can be determined to fall within the vertex enclosing region. Therefore, a correspondence relationship can be established between the i-th anchor point and the vertex enclosing region, such that the correspondence relationship indicates that the vertex enclosing region is the vertex enclosing region corresponding to the i-th anchor point in the second planar representation image. Subsequently, based on the region description information of the vertex enclosing region corresponding to the i-th anchor point in the second planar representation image, second position representation data for the i-th anchor point can be determined, such that the second position representation data can represent the characteristics of the reference three-dimensional mesh presented at the i-th anchor point. The implementation of the region description information is similar to that of the region description information above and will not be further described here for the sake of brevity. i is a positive integer, i≤I, I is a positive integer, I represents the number of anchor points in the n-th anchor point set, and n is a positive integer, n≤N, N represents the number of anchor point sets.

[0125] It can be seen that in one possible implementation, when the area enclosed by each vertex in the second planar representation image is a triangular face, for the i-th anchor point in the n-th anchor point set, if the i-th anchor point falls within a triangular face in the second planar representation image, a correspondence between the anchor point and the triangular face is established, so that the second position representation data of the i-th anchor point (e.g., the identifier of the triangular face and the centroid coordinates of the i-th anchor point within the triangular face) can be determined based on the area description information of the triangular face corresponding to the i-th anchor point in the second planar representation image (e.g., at least one of the number of the triangular face and the planar position of each vertex in the triangular face in a preset two-dimensional plane space), so that the second position representation data can represent the characteristics of the reference three-dimensional mesh at the i-th anchor point. Wherein, i is a positive integer, i≤I, I is a positive integer, I represents the number of anchor points in the n-th anchor point set, and n is a positive integer, n≤N, N represents the number of anchor point sets.

[0126] Based on the relevant content of S3 above, it can be known that for each anchor point in each anchor point set, after obtaining the first plane representation image of the three-dimensional mesh to be processed and the second plane representation image of the reference three-dimensional mesh, if it is determined that the anchor point (for example, anchor point 1 shown in Figure 5) falls within a vertex enclosed area in the first plane representation image, then it can be determined that the anchor point belongs to a valid point corresponding to the first plane representation image, and based on the area description information of the vertex enclosed area (for example, the number of the triangle face, and the UV coordinates of each vertex in the triangle face, etc.), the first position representation data of the anchor point (for example, the number of the triangle face and the centroid coordinates of the anchor point in the triangle face, etc.) can be determined, so that the first position representation data can reflect the characteristics of the three-dimensional mesh to be processed under the i-th anchor point; however, if it is determined that the anchor point (for example, anchor point 2 shown in Figure 5) does not fall within any vertex enclosed area in the first plane representation image, then it can be determined that the anchor point belongs to an invalid point corresponding to the first plane representation image. Similarly, if it is determined that the anchor point falls within a vertex enclosed area in the second plane representation image, the second position representation data of the anchor point (for example, the number of the triangle facet and the centroid coordinates of the anchor point within the triangle facet, etc.) can be determined based on the area description information of the vertex enclosed area (for example, the number of the triangle facet and the UV coordinates of each vertex in the triangle facet, etc.), so that the second position representation data can reflect the characteristics of the reference three-dimensional grid under the i-th anchor point; however, if it is determined that the anchor point does not fall within any vertex enclosed area in the second plane representation image, it can be determined that the anchor point belongs to an invalid point corresponding to the second plane representation image.

[0127] It should be noted that since the three-dimensional grid near the eyes and oral cavity is relatively complex, in some application scenarios, the anchor points falling into the eye area or the inner area of ​​the oral cavity (for example, anchor point 3 shown in Figure 5) can be directly regarded as invalid points.

[0128] S4: Determine mesh performance representation data of the three-dimensional mesh to be processed based on the difference representation data between the first position representation data and the second position representation data.

[0129] Among them, the grid performance characterization data of the three-dimensional grid to be processed refers to the grid performance quantification result of the three-dimensional grid to be processed, so that the grid performance characterization data is used to represent the state (e.g., quantitative results, etc.) of the three-dimensional grid to be processed under one or more grid performance evaluation dimensions (e.g., standardization, etc.).

[0130] In addition, the present disclosure does not limit the implementation method of the mesh performance characterization data of the three-dimensional mesh to be processed above. For example, in some application scenarios (for example, a quantitative processing scenario for the standardization of a reconstructed three-dimensional mesh), the mesh performance characterization data can be used to represent the quantitative results presented by the three-dimensional mesh to be processed under standardization, so that the mesh performance characterization data can represent the degree of standardization of the three-dimensional mesh to be processed. It should be noted that the present disclosure does not limit the correlation between the mesh performance characterization data and the degree of standardization. For example, in some application scenarios, the mesh performance characterization data is negatively correlated with the degree of standardization, that is, if the mesh performance characterization data is smaller, it can indicate that the degree of standardization of the three-dimensional mesh to be processed is higher (that is, more standardized), and vice versa, it indicates that the degree of standardization of the three-dimensional mesh to be processed is lower (that is, more severely distorted).

[0131] In addition, the present disclosure does not limit the implementation of the above S4. For example, in some application scenarios, when the above preset two-dimensional plane space includes N anchor point sets, the S4 can be specifically as follows: first, for the i-th anchor point in the n-th anchor point set (for example, anchor point 1 shown in Figure 5), after obtaining the first position representation data and the second position representation data of the i-th anchor point, the difference representation data between the first position representation data and the second position representation data can be calculated, so that the difference representation data can represent the difference between the three-dimensional grid to be processed and the reference three-dimensional grid at the i-th anchor point. Heterogeneity, i is a positive integer, i≤I, I is a positive integer, I represents the number of anchor points in the n-th anchor point set, n is a positive integer, n≤N, N represents the number of anchor point sets; then, based on the difference characterization data corresponding to all anchor points in all anchor point sets, the grid performance characterization data of the three-dimensional grid to be processed is determined, so that the grid performance characterization data can represent the difference presented by the three-dimensional grid to be processed relative to the reference three-dimensional grid, so that the grid performance characterization data can represent the state presented by the three-dimensional grid to be processed under one or more grid performance evaluation dimensions (such as standardization, etc.).

[0132] It should be noted that the present disclosure does not limit the implementation method of the step of "determining the grid performance characterization data of the three-dimensional grid to be processed based on the difference characterization data corresponding to all anchor points in all anchor point sets" in the previous paragraph. For example, it can be specifically: directly adding up the difference characterization data corresponding to all anchor points in all anchor point sets to obtain the grid performance characterization data of the three-dimensional grid to be processed.

[0133] In fact, in order to better improve the evaluation effect, the present disclosure also provides a possible implementation method of the process of determining the grid performance characterization data of the three-dimensional grid to be processed above. Under this implementation method, the process of determining the grid performance characterization data of the three-dimensional grid to be processed may include the following steps 21-22.

[0134] Step 21: For any anchor point set, a first virtual grid and a second virtual grid are constructed using the anchor point set, and a grid performance evaluation result corresponding to the anchor point set is determined based on the grid difference representation data between the first virtual grid and the second virtual grid; the vertices in the first virtual grid are determined based on the first anchor point in the anchor point set, and there is a vertex enclosing area corresponding to the first anchor point in the first plane representation image, and the first description information of the vertices in the first virtual grid is determined based on the first position representation data of the first anchor point; the vertices in the second virtual grid are determined based on the second anchor point in the anchor point set, and there is a vertex enclosing area corresponding to the second anchor point in the second plane representation image, and the second description information of the vertices in the second virtual grid is determined based on the second position representation data of the second anchor point; the grid difference representation data is determined based on the first description information of the vertices in the first virtual grid and the second description information of the vertices in the second virtual grid.

[0135] In the present disclosure, for the nth anchor point set, the first virtual grid constructed using the nth anchor point set can be used to represent the state of the above-mentioned three-dimensional grid to be processed under the nth anchor point set, so that the first virtual grid can represent the characteristics of the above-mentioned three-dimensional grid to be processed under the nth anchor point set. The vertices in the first virtual grid are determined based on the first anchor point in the nth anchor point set, and the first description information of the vertices in the first virtual grid is determined based on the first position representation data of the first anchor point. The first anchor point refers to the valid point corresponding to the first plane representation image in the nth anchor point set, so that the vertex enclosing area corresponding to the first anchor point exists in the first plane representation image. That is, for the nth anchor point set, the first anchor point in the nth anchor point set can refer to an anchor point in the nth anchor point set that falls within a vertex inclusion area in the first plane representation image. In addition, for any vertex in the first virtual grid, if the vertex is determined based on a first anchor point in the nth anchor point set, the first description information of the vertex is determined based on the first position representation data of the first anchor point (for example, the first position representation data of the first anchor point is directly determined as the first description information of the vertex, etc.). The first description information is used to describe the state of the vertex. In addition, because the first position representation data of the first anchor point can characterize the characteristics of the three-dimensional grid to be processed under this first anchor point, the first description information can also characterize the characteristics of the three-dimensional grid to be processed under this first anchor point, so that the first virtual grid including the vertex can characterize the characteristics of the three-dimensional grid to be processed under this first anchor point. n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0136] Similarly, for the nth anchor point set, the second virtual mesh constructed using the nth anchor point set can be used to represent the state of the reference three-dimensional mesh under the nth anchor point set, so that the second virtual mesh can represent the characteristics of the reference three-dimensional mesh under the nth anchor point set. The vertices in the second virtual mesh are determined based on the second anchor point in the nth anchor point set, and the second description information of the vertices in the second virtual mesh is determined based on the second position representation data of the second anchor point. The second anchor point refers to a valid point corresponding to the second plane representation image in the nth anchor point set, so that the second plane representation image contains a vertex enclosed area corresponding to the second anchor point. That is, for the nth anchor point set, the second anchor point in the nth anchor point set can refer to an anchor point in the nth anchor point set that falls within a vertex enclosed area in the second plane representation image. In addition, for any vertex in the second virtual grid, if the vertex is determined based on a second anchor point in the nth anchor point set, the second description information of the vertex is determined based on the second position representation data of the second anchor point (for example, the second position representation data of the second anchor point is directly determined as the second description information of the vertex, etc.). The second description information is used to describe the state of the vertex. In addition, because the second position representation data of the second anchor point can characterize the characteristics of the reference three-dimensional grid under this second anchor point, the second description information can also characterize the characteristics of the reference three-dimensional grid under this second anchor point, so that the second virtual grid including the vertex can characterize the characteristics of the reference three-dimensional grid under this second anchor point. n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0137] In fact, in order to better improve the grid performance evaluation effect, the present disclosure also provides an implementation method of the above step of "using the anchor point set to construct the first virtual grid and the second virtual grid", which may specifically include the following steps 211 to 213.

[0138] Step 211: Determine each anchor point group to be used from the nth anchor point set; for any anchor point group to be used, the rectangle formed by all anchor points in the anchor point group to be used satisfies a preset rectangle condition, where n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0139] The anchor point group to be used includes some anchor points in the nth anchor point set, and the rectangle formed by these anchor points satisfies a preset rectangle condition. The preset rectangle condition refers to a condition required to be based on when determining the anchor point group to be used from the nth anchor point set; and the preset rectangle condition can be set according to the actual application scenario. For example, in some application scenarios, the preset rectangle condition can specifically be: a minimum rectangle.

[0140] It can be seen that in one possible implementation, for the nth anchor point set, an anchor point group that can form a minimum rectangle (for example, the anchor point group including anchor point a1, anchor point a2, anchor point a3 and anchor point a4 shown in Figure 6) can be searched from the nth anchor point set, and each anchor point group that can form a minimum rectangle is used as the anchor point group to be used, so that the anchor point group to be used can represent the anchor point group that exists in the nth anchor point set and can form a minimum rectangle.

[0141] Based on the relevant content of step 211 above, it can be seen that for the nth anchor point set, some anchor point groups to be used can be found from the nth anchor point set, so that the rectangles formed by all anchor points in each anchor point group to be used meet the preset rectangle conditions, so that the virtual three-dimensional mesh can be constructed using these anchor point groups to be used. Where n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0142] Step 212: For any anchor point group to be used determined from the nth anchor point set, if there exists a vertex enclosing region corresponding to each anchor point in the anchor point group to be used in the first planar representation image, a first virtual grid is constructed based on all anchor points in the anchor point group to be used and the first position representation data of all anchor points; if there exists a vertex enclosing region corresponding to some anchor points in the anchor point group to be used in the first planar representation image, and the number of these anchor points is not less than a preset number threshold, a first virtual grid is constructed based on these anchor points and the first position representation data of these anchor points. Where n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0143] In the present disclosure, for the gth anchor point group to be used determined from the nth anchor point set, if there is a vertex enclosing area corresponding to each anchor point in the gth anchor point group to be used in the first plane representation image, it can be determined that all anchor points in the gth anchor point group to be used fall into the corresponding vertex enclosing area in the first plane representation image, and thus it can be determined that all anchor points in the gth anchor point group to be used belong to valid points corresponding to the first plane representation image. Therefore, a first virtual grid can be directly constructed based on all anchor points in the gth anchor point group to be used and the first position representation data of all anchor points, so that the first virtual grid includes all anchor points in the gth anchor point group to be used, the closed area enclosed by all anchor points (for example, two triangles), and the first position representation data of all anchor points. Wherein, g is a positive integer, g≤G, G is a positive integer, and G represents the number of anchor point groups to be used determined from the nth anchor point set.

[0144] It should be noted that the present disclosure does not limit the implementation method of the step of "constructing a first virtual grid based on all anchor points in the g-th anchor point group to be used" in the previous paragraph. For example, in some application scenarios, when the g-th anchor point group to be used includes four anchor points, it can specifically be: based on the four anchor points in the g-th anchor point group to be used (for example, the four anchor points of anchor point a1, anchor point a2, anchor point a3 and anchor point a4 shown in Figure 6) and the first position representation data of the four anchor points, construct two triangular faces in the first virtual grid (for example, the two triangular faces shown in Figure 6), so that the two triangular faces can represent the connection relationship between the four anchor points and the first position representation data of the four anchor points.

[0145] It should also be noted that the present disclosure does not limit the implementation method of the step of "constructing two triangular faces in the first virtual grid based on the four anchor points in the g-th anchor point group to be used and the first position representation data of the four anchor points" in the previous paragraph. For example, it can be specifically as follows: first connect the four anchor points in the g-th anchor point group to be used to obtain a rectangle (for example, the minimum rectangle surrounded by the four anchor points), so that the four vertices of the rectangle are the four anchor points, and the first description information of the four vertices of the rectangle are the four anchor points; then divide the rectangle into two triangular faces in a preset manner to obtain two triangular faces in the first virtual grid. The preset manner can be pre-set, for example, it can be implemented by any existing or future method that can divide a rectangle into two triangles (for example, the method of adding a diagonal line to the rectangle as shown in FIG6 ).

[0146] In addition, for the g-th anchor point group to be used determined from the n-th anchor point set, if there is a vertex enclosing area corresponding to some anchor points (for example, one anchor point, two anchor points or three anchor points) in the g-th anchor point group to be used in the first plane representation image above, it can be determined that some anchor points in the g-th anchor point group to be used fall into the corresponding vertex enclosing area in the first plane representation image, so that it can be determined that only some anchor points in the g-th anchor point group to be used belong to the valid points corresponding to the first plane representation image, so it can be further determined whether the number of the some anchor points is lower than a preset number threshold (for example, 3), so as to determine that the number of the some anchor points is not lower than the preset number threshold (for example, 3 anchor points in the g-th anchor point group to be used belong to the valid points corresponding to the first plane representation image). When the number of the anchor points is less than a preset number threshold (for example, two or less anchor points in the g-th anchor point group to be used belong to valid points corresponding to the first plane representation image, etc.), it can be determined that the anchor points can constitute a closed area in the first virtual grid, so the first virtual grid can be constructed based on the anchor points and the first position representation data of the anchor points, so that the first virtual grid includes the anchor points, the closed area formed by the anchor points, and the first position representation data of the anchor points; however, if it is determined that the number of the anchor points is less than a preset number threshold (for example, two or less anchor points in the g-th anchor point group to be used belong to valid points corresponding to the first plane representation image, etc.), it can be determined that the anchor points cannot constitute a closed area in the first virtual grid, so the g-th anchor point group to be used can be directly discarded. The preset number threshold refers to the minimum number of vertices required to construct a closed area, g is a positive integer, g≤G, G is a positive integer, and G represents the number of anchor point groups to be used determined from the n-th anchor point set.

[0147] It should be noted that the present disclosure does not limit the implementation method of the step of "constructing a first virtual grid based on the part of anchor points and the first position representation data of the part of anchor points" in the previous paragraph. For example, in some application scenarios, when the g-th anchor point group to be used includes four anchor points, and three of the four anchor points belong to valid points corresponding to the first plane representation image, it can be specifically: based on the three anchor points and the first position representation data of the three anchor points, construct a triangular face in the first virtual grid, so that this triangular face can represent the connection relationship between the three anchor points and the first position representation data of the three anchor points.

[0148] In addition, for the g-th anchor point group to be used determined from the n-th anchor point set, if there is no vertex enclosing area corresponding to each anchor point in the g-th anchor point group to be used in the first plane representation image, it can be determined that all anchor points in the g-th anchor point group to be used do not fall into the corresponding vertex enclosing area in the first plane representation image, and thus it can be determined that all anchor points in the g-th anchor point group to be used do not belong to the valid points corresponding to the first plane representation image, so the g-th anchor point group to be used can be directly discarded. Wherein, g is a positive integer, g≤G, G is a positive integer, and G represents the number of anchor point groups to be used determined from the n-th anchor point set.

[0149] Based on the relevant content of step 212 above, it can be known that for the nth anchor point set, after finding the gth anchor point group to be used that can form the minimum rectangle from the nth anchor point set, if it is determined that all anchor points in the gth anchor point group to be used belong to valid points corresponding to the first plane representation image, the minimum rectangle formed by the gth anchor point group to be used can be divided into two triangles, and both triangles are used as closed areas in the first virtual grid; if it is determined that 3 anchor points in the gth anchor point group to be used belong to the first plane representation image, If the g-th anchor point group is a valid point corresponding to the first plane representation image, the three anchor points can be connected into a triangle, and the triangle can be used as a closed area in the first virtual grid; if it is determined that two anchor points in the g-th anchor point group to be used belong to the valid points corresponding to the first plane representation image, or if it is determined that one anchor point in the g-th anchor point group to be used belongs to the valid point corresponding to the first plane representation image, or if it is determined that zero anchor points in the g-th anchor point group to be used belong to the valid points corresponding to the first plane representation image, then the g-th anchor point group to be used can be directly discarded. Wherein, g is a positive integer, g≤G, G is a positive integer, G represents the number of anchor point groups to be used determined from the n-th anchor point set, n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0150] Step 213: For any anchor point group to be used determined from the nth anchor point set, if the second planar representation image contains vertex enclosed areas corresponding to each anchor point in the anchor point group to be used, then a second virtual grid is constructed based on all anchor points in the anchor point group to be used and the second position representation data of all anchor points; if the second planar representation image contains vertex enclosed areas corresponding to some anchor points in the anchor point group to be used, and the number of these anchor points is not less than a preset number threshold, then a second virtual grid is constructed based on these anchor points and the second position representation data of these anchor points. Where n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0151] It should be noted that the implementation of step 213 is similar to the implementation of step 212 above, and for the sake of brevity, it will not be repeated here.

[0152] It can be seen that in a possible embodiment, the above step 213 can be specifically as follows: for any anchor point group to be used determined from the nth anchor point set, when the anchor point group to be used includes four anchor points, if there is a vertex enclosing area corresponding to each anchor point in the anchor point group to be used in the second plane representation image, it can be determined that the four anchor points in the anchor point group to be used are valid points corresponding to the second plane representation image, and therefore two triangular faces in the second virtual mesh can be constructed based on the four anchor points in the anchor point group to be used and the second position representation data of the four anchor points, so that the two triangular faces can represent the connection relationship between the four anchor points and the second position representation data of the four anchor points; if there is a vertex enclosing area corresponding to three anchor points in the anchor point group to be used in the second plane representation image, it can be determined that the three anchor points in the anchor point group to be used are valid points corresponding to the second plane representation image, and therefore one triangular face in the second virtual mesh can be constructed based on the three anchor points and the second position representation data of the three anchor points. Wherein, n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0153] Based on the relevant contents of steps 211 to 213 above, it can be known that in some application scenarios, for the nth anchor point set, a first virtual grid (for example, the virtual grids shown in FIG7 ) can be constructed based on the first anchor point in the nth anchor point set (that is, the valid point corresponding to the first plane representation image), so that the first virtual grid includes some or all of the valid points corresponding to the first plane representation image in the nth anchor point set, the first position representation data of these valid points, and the connection relationship between these valid points, so that the first virtual grid can represent the position of the three-dimensional grid to be processed under the nth anchor point set. Similarly, a second virtual grid can be constructed based on the second anchor point in the nth anchor point set (i.e., the valid point corresponding to the second plane representation image), so that the second virtual grid includes some or all of the valid points corresponding to the second plane representation image in the nth anchor point set, the second position representation data of these valid points, and the connection relationship between these valid points, so that the second virtual grid can represent the state presented by the reference three-dimensional grid under the nth anchor point set, so that the grid performance of the three-dimensional grid to be processed can be evaluated subsequently by using the difference between the first virtual grid and the second virtual grid. Wherein, n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0154] It can be seen that for the nth anchor point set, after the first virtual grid and the second virtual grid are constructed using the nth anchor point set, the grid difference characterization data between the first virtual grid and the second virtual grid can be calculated based on the first description information of the vertices in the first virtual grid and the second description information of the vertices in the second virtual grid, so that the grid difference characterization data can represent the difference between the first virtual grid and the second virtual grid; and then based on the grid difference characterization data, the grid performance evaluation result corresponding to the nth anchor point set is determined, so that the grid performance evaluation result can represent the grid performance of the three-dimensional grid to be processed relative to the reference three-dimensional network on the nth anchor point set. Among them, the grid performance evaluation result corresponding to the nth anchor point set is used to represent the grid performance of the three-dimensional grid to be processed on the nth anchor point set. n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0155] It should be noted that the present disclosure does not limit the method for determining the above “grid difference characterization data between the first virtual grid and the second virtual grid”. For example, it can be implemented with the help of the following formula (8).

[0156] Where ΔX n represents the grid difference characterization data between the first virtual grid and the second virtual grid, and the ΔX n It may include vertex difference representation data of several vertices (that is, the difference between the first description information of a vertex and the second description information of the vertex); n represents the first virtual grid, and the X n The first description information of these vertices may be included; represents the second virtual grid, and the Second description information for these vertices may be included.

[0157] It should also be noted that the present disclosure does not limit the process of determining the grid performance evaluation result corresponding to the nth anchor point set above. For example, it can be implemented with the help of the Laplace loss shown in the following formulas (9)-(12).

[0158] L n =∑ p δ p (9)

[0159] δ p =∑ {p,q}∈Ε w pq (v q -v p )=[∑ {p,q}∈Ε w pq vq ]-v p (10)

[0160] ω pq =cotα+cotβ (12)

[0161] Where, L n represents the grid performance evaluation result corresponding to the nth anchor point set above; δ p represents the Laplace loss corresponding to the p-th vertex in the grid difference representation data between the first virtual grid and the second virtual grid; ∑ p δ p Indicates that the Laplace losses corresponding to all vertices in the mesh difference representation data are summed; v p represents the vertex difference representation data of the p-th vertex; v q represents the vertex difference representation data of the qth adjacent vertex that has a connection relationship with the pth vertex; {p,q}∈Ε indicates that there is an edge between the pth vertex and the qth adjacent vertex, that is, there is a connection relationship; w pq Represents an edge <v p ,v q >The corresponding normalized weights, and ∑ {p,q}∈Ε w pq =1; α and β are the edges <v p ,v q >The two objects in the triangle (for example, α and β shown in Figure 8).

[0162] Based on the relevant content of the Laplace loss above, it can be known that in some application scenarios, for the first virtual grid and the second virtual grid constructed using the nth anchor point set, the grid difference representation data between the first virtual grid and the second virtual grid may include vertex difference representation data of several vertices. Among them, for any vertex (for example, the pth vertex above), the vertex difference representation data of the vertex (for example, the vth vertex above) p ) is determined based on the difference between the first description information of the vertex and the second description information of the vertex (for example, determined using the above formula (8)); and the mesh performance evaluation result corresponding to the n-th anchor point set can be based on the Laplace loss corresponding to several vertices (for example, the above δ p ), for any vertex, the Laplace loss corresponding to the vertex is based on the vertex difference representation data of the vertex (for example, the above v p ), the vertex difference representation data of the adjacent vertices of the vertex (for example, the above v q), determined by the two diagonal angles (for example, α and β above) corresponding to the edge formed by the vertex and its adjacent vertices.

[0163] It should be noted that for the qth vertex above, if a vertex is connected to the qth vertex, then the vertex can be considered a neighbor of the qth vertex. q is a positive integer, q≤Q, where Q represents the number of vertices involved in the "mesh difference representation data between the first virtual mesh and the second virtual mesh" above.

[0164] Based on the relevant content of step 21 above, it can be known that for the nth anchor point set, after obtaining the first position representation data and the second position representation data of the anchor points in the nth anchor point set, a first virtual grid can be first constructed based on the first anchor point in the nth anchor point set and the first position representation data of the first anchor point, and a second virtual grid can be constructed based on the second anchor point in the nth anchor point set and the second position representation data of the second anchor point; then, grid difference representation data between the first virtual grid and the second virtual grid is calculated; and then, a grid performance evaluation result corresponding to the nth anchor point set is determined based on the grid difference representation data, so that the grid performance evaluation result can characterize the grid performance of the three-dimensional grid to be processed on the nth anchor point set. Wherein, n is a positive integer, n≤N, and N represents the number of anchor point sets.

[0165] Step 22: Determine mesh performance characterization data of the three-dimensional mesh to be processed based on the mesh performance evaluation result corresponding to at least one anchor point set.

[0166] It should be noted that the present disclosure does not limit the implementation of step 22. For example, in some possible implementations, step 22 may specifically include taking the average of the mesh performance evaluation results corresponding to all anchor point sets as the mesh performance representation data for the three-dimensional mesh to be processed. For another example, in some possible implementations, step 22 may specifically include combining the mesh performance evaluation results corresponding to all anchor point sets to obtain the mesh performance representation data for the three-dimensional mesh to be processed.

[0167] Based on the relevant content of steps 21 to 22 above, it can be known that in some application scenarios, the grid performance characterization data of the above-mentioned three-dimensional grid to be processed can be determined by constructing several groups of virtual three-dimensional grids covering the entire preset two-dimensional plane space, so that the grid performance characterization data can accurately, comprehensively, and fine-grainedly represent the quantitative results presented by the three-dimensional grid to be processed under one or more grid performance evaluation dimensions (such as standardization, etc.), which is conducive to improving the grid performance evaluation effect. In addition, the present disclosure also uses Laplace loss to determine the grid performance characterization data of the three-dimensional grid to be processed, so that the grid performance characterization data can more accurately, comprehensively, and fine-grainedly represent the quantitative results presented by the three-dimensional grid to be processed under one or more grid performance evaluation dimensions (such as standardization, etc.), which is conducive to better improving the grid performance evaluation effect.

[0168] Based on the relevant contents of S1 to S4 above, it can be known that for the data processing method provided by the embodiment of the present disclosure, after obtaining the three-dimensional grid to be processed (for example, the reconstructed three-dimensional grid) and the reference three-dimensional grid (for example, a pre-set normalized three-dimensional grid template), a first plane representation image of the three-dimensional grid to be processed and a second plane representation image of the reference three-dimensional grid are drawn in a preset two-dimensional plane space (for example, a plane blank image), so that the first plane representation image is used to represent the state of the vertices in the three-dimensional grid to be processed in the preset two-dimensional plane space (for example, the position distribution state of the vertices, and the connection state between different vertices, etc.), and the second plane representation image is used to represent the state of the vertices in the reference three-dimensional grid in the preset two-dimensional plane space, so that when the preset two-dimensional plane When the space includes at least one anchor point set, the first position representation data and the second position representation data of each anchor point in the anchor point set are determined, wherein the first position representation data is determined based on the area description information of the vertex enclosed area corresponding to the anchor point in the first plane representation image, so that the first position representation data can, to a certain extent, represent the state of the three-dimensional mesh to be processed on the anchor point; the second position representation data is determined based on the area description information of the vertex enclosed area corresponding to the anchor point in the second plane representation image, so that the second position representation data can, to a certain extent, represent the state of the reference three-dimensional mesh on the anchor point; then, based on the difference representation data between the first position representation data and the second position representation data, the mesh performance representation data of the three-dimensional mesh to be processed is determined. Among them, because the difference characterization data can represent the difference between the three-dimensional grid to be processed and the reference three-dimensional grid, the grid performance characterization data determined based on the difference characterization data can represent the grid performance presented by the three-dimensional grid to be processed relative to the reference three-dimensional grid, so that the grid performance characterization data can represent the grid performance quantification results presented by the three-dimensional grid to be processed under one or more grid performance evaluation dimensions (for example, dimensions such as standardization). In this way, the purpose of quantitative evaluation processing of a three-dimensional grid under these grid performance evaluation dimensions can be achieved, thereby effectively avoiding defects (for example, poor grid performance evaluation effect, etc.) caused by the inability to quantify a certain grid performance evaluation dimension (for example, dimensions such as standardization) or the poor quantification scheme for a certain grid performance evaluation dimension (for example, dimensions such as accuracy), which is conducive to improving the grid performance evaluation effect.

[0169] In addition, the present disclosure does not limit the application scenarios of the above data processing method. For ease of understanding, two application scenarios are used as examples for illustration below.

[0170] In scenario 1, the data processing method provided in this disclosure can be used to implement the model training process.

[0171] Based on this, it can be seen that in one possible implementation, when the above-mentioned three-dimensional mesh to be processed is constructed using a three-dimensional reconstruction model, the data processing method provided by the present disclosure includes not only S1-S4 above, but also the following step 31. The execution time of step 31 is later than the execution time of S4.

[0172] Step 31: Update the 3D reconstruction model based on the mesh performance characterization data of the 3D mesh to be processed.

[0173] The three-dimensional reconstruction model is used to perform three-dimensional reconstruction processing on input data of the three-dimensional reconstruction model (for example, an image).

[0174] Based on the relevant content of step 31 above, it can be known that for the three-dimensional reconstruction model, after obtaining the three-dimensional grid to be processed constructed by the three-dimensional reconstruction model, the above-mentioned reference three-dimensional grid can be used as label information to perform grid performance evaluation processing on the three-dimensional grid to be processed to obtain grid performance characterization data of the three-dimensional grid to be processed, so that the grid performance characterization data can reflect the three-dimensional reconstruction performance of the three-dimensional reconstruction model to a certain extent, so that the three-dimensional reconstruction model can be subsequently updated based on the grid performance characterization data, so that the updated three-dimensional reconstruction model has better three-dimensional reconstruction performance, so that the new three-dimensional grid constructed using the updated three-dimensional reconstruction model presents a better performance status under one or more grid performance evaluation dimensions (for example, standardization, etc.).

[0175] It can be seen that, in a possible implementation, the training process of the above three-dimensional reconstruction model may include the following steps 32 to 37.

[0176] Step 32: Determine an image to be processed from the sample image set.

[0177] The sample image set refers to a training data set required for training a three-dimensional reconstruction model; and the present disclosure does not limit the implementation method of the sample image set.

[0178] The image to be processed refers to an image extracted from the sample image set above and requiring 3D reconstruction processing in the current round of training; and the image to be processed is used to describe the object to be processed. The object to be processed refers to the object described by the image to be processed.

[0179] In addition, the present disclosure does not limit the method of obtaining the above-mentioned images to be processed. For example, for any round of training process, the process of obtaining the images to be processed can be: randomly selecting images from the images in the sample image set that have not been processed, and using the selected images as the images to be processed.

[0180] In addition, the present disclosure does not limit the number of the above-mentioned images to be processed.

[0181] Based on the relevant content of step 32 above, it can be known that in some application scenarios, for the current round of training process, the image to be processed can be determined from the sample image set so that the model performance evaluation and model update processing can be performed based on the image to be processed.

[0182] Step 33: Perform three-dimensional reconstruction on the image to be processed using the three-dimensional reconstruction model to obtain a three-dimensional mesh to be processed corresponding to the image to be processed, so that the three-dimensional mesh to be processed is used to describe the state of the object to be processed in the three-dimensional space.

[0183] In the present disclosure, for the current round of training process, after obtaining the image to be processed, the image to be processed can be input into the three-dimensional reconstruction model so that the three-dimensional reconstruction model can perform three-dimensional reconstruction processing on the image to be processed, obtain and output the three-dimensional grid to be processed corresponding to the image to be processed, so that the processed three-dimensional grid is used to describe the state of the object to be processed in the three-dimensional space, so that the model performance evaluation processing can be carried out subsequently based on the three-dimensional grid to be processed and the reference three-dimensional grid above.

[0184] It should be noted that the present disclosure does not limit the implementation method of the reference three-dimensional grid. For example, in some application scenarios (for example, scenarios where it is necessary to perform quantitative processing on the standardization of the three-dimensional grid to be processed, etc.), the reference three-dimensional grid may refer to a pre-set normalized three-dimensional grid template, and the reference three-dimensional grid may remain unchanged during multiple rounds of training, so that there is no need to obtain a reference three-dimensional grid once in each round of training, which is conducive to simplifying the model training process. For another example, in some application scenarios (for example, scenarios where it is necessary to perform quantitative processing on the accuracy of the three-dimensional grid to be processed, etc.), the reference three-dimensional grid may refer to a label three-dimensional grid set in advance for the above-mentioned image to be processed, so that the label three-dimensional grid can accurately represent the state of the object to be processed in the three-dimensional space, so that the reference three-dimensional grid will be updated accordingly as the image to be processed is updated, which is conducive to improving the model training effect.

[0185] Step 34: Draw a first plane representation image of the three-dimensional mesh to be processed and a second plane representation image of the reference three-dimensional mesh in a preset two-dimensional plane space; the first plane representation image is used to represent the state of the vertices in the three-dimensional mesh to be processed in the preset two-dimensional plane space; the second plane representation image is used to represent the state of the vertices in the reference three-dimensional mesh in the preset two-dimensional plane space; the preset two-dimensional plane space includes at least one anchor point set.

[0186] It should be noted that the relevant content of step 34 can be found in S2 above, and for the sake of brevity, it will not be repeated here.

[0187] Step 35: Determine the first position representation data and the second position representation data of the anchor point in each anchor point set; the first position representation data is determined based on the area description information of the vertex enclosed area corresponding to the anchor point in the first plane representation image; the second position representation data is determined based on the area description information of the vertex enclosed area corresponding to the anchor point in the second plane representation image.

[0188] It should be noted that the relevant content of step 35 can be found in S3 above, and for the sake of brevity, it will not be repeated here.

[0189] Step 36: Determine mesh performance representation data of the to-be-processed three-dimensional mesh based on the difference representation data between the first position representation data and the second position representation data.

[0190] It should be noted that the relevant content of step 36 can be found in S4 above, and for the sake of brevity, it will not be repeated here.

[0191] Step 37: Update the 3D reconstruction model based on the mesh performance characterization data of the 3D mesh to be processed, and return to execute the above step 32 and subsequent steps until a preset stop condition is reached.

[0192] The preset stopping condition refers to a condition that must be met to terminate the model training process; and the present disclosure does not limit the preset stopping condition. For example, the preset stopping condition may include: the model loss of the 3D reconstructed model is lower than a preset loss threshold. In another example, the preset stopping condition may include: the rate of change of the model loss of the 3D reconstructed model is lower than a preset rate of change threshold. In another example, the preset stopping condition may include: the number of updates to the 3D reconstructed model reaches a preset number threshold.

[0193] Furthermore, the model loss of the 3D reconstructed model in the previous section is used to characterize the 3D reconstruction performance of the 3D reconstructed model; this model loss is determined based on the mesh performance characterization data of the 3D mesh to be processed. It should be noted that this disclosure does not limit the method for obtaining this model loss.

[0194] Based on the relevant content of steps 32 to 37 above, it can be seen that in some application scenarios, for a 3D reconstruction model, the 3D reconstruction model can complete 3D reconstruction learning with the help of multiple rounds of training processes, so that the finally trained 3D reconstruction model can have better 3D reconstruction performance, so that the 3D reconstruction model can be used to complete some 3D reconstruction tasks in the future (for example, the rendering processing tasks involved in scenario 2 below).

[0195] Scenario 2: The data processing method provided in this disclosure can be used to implement a rendering processing scenario.

[0196] Based on this, it can be seen that in one possible implementation, when the above-mentioned 3D mesh to be processed is obtained by performing 3D reconstruction processing on the image provided by the user, the data processing method provided by the present disclosure includes not only S1-S4 above, but also steps 41-42 below. The execution time of step 41 is later than the execution time of step S4.

[0197] Step 41: If the mesh performance characterization data of the three-dimensional mesh to be processed reaches a preset performance threshold, the target map is rendered to the three-dimensional mesh to be processed to obtain a rendering result.

[0198] The preset performance threshold is used to indicate the minimum state that a three-dimensional grid needs to reach in terms of grid performance in order to be able to perform texture rendering processing; and the preset performance threshold can be set according to an actual application scenario.

[0199] The target texture refers to a texture material pre-specified by the user, so that the target texture represents the texture required for rendering the 3D mesh to be processed. It should be noted that this disclosure does not limit the implementation of the target texture. For example, in some application scenarios, the target texture may refer to a texture material created by relevant personnel based on a pre-defined 3D mesh template and UV unwrapped diagram.

[0200] The rendering result refers to the result obtained by using the target map to perform map rendering processing on the processed three-dimensional mesh; and the present disclosure does not limit the method of obtaining the rendering result. For example, it can be implemented using any existing or future method that can render the map to the three-dimensional mesh (for example, a rendering processing method implemented by a pre-built three-dimensional rendering engine, etc.).

[0201] Step 42: Display the rendering result.

[0202] Based on the relevant content of steps 41 to 42 above, it can be known that for some application scenarios, after obtaining the image provided by the user, a three-dimensional reconstruction process can be first performed on the image to obtain a three-dimensional mesh to be processed; then, based on the reference three-dimensional mesh in the above text, a mesh performance evaluation process is performed on the three-dimensional mesh to be processed to obtain mesh performance characterization data of the three-dimensional mesh to be processed, so that the mesh performance characterization data can represent the state of the three-dimensional mesh to be processed under one or more mesh performance evaluation dimensions (such as standardization, etc.), so that when it is determined that the mesh performance characterization data reaches a preset performance threshold, it can be determined that the three-dimensional mesh to be processed has good mesh performance, so the target map specified by the user can be rendered to the three-dimensional mesh to be processed and the rendering result can be displayed so that the user can see the rendering result. Among them, because the three-dimensional mesh to be processed has good mesh performance, the final rendering result can better meet the image processing requirements, thereby helping to improve the image processing effect.

[0203] Based on the data processing method provided in the embodiments of the present disclosure, the embodiments of the present disclosure also provide a data processing device, which will be explained and illustrated below in conjunction with Figure 9. Figure 9 is a schematic diagram of the structure of a data processing device provided in the embodiments of the present disclosure. It should be noted that for the technical details of the data processing device provided in the embodiments of the present disclosure, please refer to the relevant content of the data processing method above.

[0204] As shown in FIG9 , the data processing device 900 provided in an embodiment of the present disclosure includes:

[0205] A grid acquisition unit 901 is used to acquire a 3D grid to be processed and a reference 3D grid;

[0206] An image drawing unit 902 is configured to draw a first planar representation image of the 3D mesh to be processed and a second planar representation image of the reference 3D mesh in a preset 2D plane space; the first planar representation image is used to represent the states of vertices in the 3D mesh to be processed in the preset 2D plane space; the second planar representation image is used to represent the states of vertices in the reference 3D mesh in the preset 2D plane space; the preset 2D plane space includes at least one anchor point set;

[0207] A first determining unit 903 is configured to determine first position representation data and second position representation data of an anchor point in each anchor point set; the first position representation data is determined based on region description information of an area enclosed by vertices corresponding to the anchor point in the first plane representation image; and the second position representation data is determined based on region description information of an area enclosed by vertices corresponding to the anchor point in the second plane representation image.

[0208] The second determining unit 904 is configured to determine mesh performance representation data of the to-be-processed three-dimensional mesh according to difference representation data between the first position representation data and the second position representation data.

[0209] In a possible implementation manner, the second determining unit 904 includes:

[0210] a first determining subunit configured to construct, for any anchor point set, a first virtual mesh and a second virtual mesh using the anchor point set, and determine a mesh performance evaluation result corresponding to the anchor point set based on mesh difference representation data between the first virtual mesh and the second virtual mesh; wherein vertices in the first virtual mesh are determined based on a first anchor point in the anchor point set, a vertex enclosing area corresponding to the first anchor point exists in the first plane representation image, and first description information of the vertices in the first virtual mesh is determined based on the first position representation data of the first anchor point; wherein vertices in the second virtual mesh are determined based on a second anchor point in the anchor point set, a vertex enclosing area corresponding to the second anchor point exists in the second plane representation image, and second description information of the vertices in the second virtual mesh is determined based on the second position representation data of the second anchor point; and wherein the mesh difference representation data is determined based on difference representation data between the first description information of the vertices in the first virtual mesh and the second description information of the vertices in the second virtual mesh;

[0211] The second determining subunit is configured to determine mesh performance characterization data of the three-dimensional mesh to be processed according to a mesh performance evaluation result corresponding to the at least one anchor point set.

[0212] In a possible implementation manner, the first determining subunit includes:

[0213] A third determining subunit is configured to determine each anchor point group to be used from the anchor point set; for any anchor point group to be used, a rectangle formed by all anchor points in the anchor point group to be used satisfies a preset rectangle condition;

[0214] a first construction subunit configured to, for any anchor point group to be used, if there exists in the first planar representation image a vertex enclosing region corresponding to each anchor point in the anchor point group to be used, construct the first virtual grid based on all anchor points in the anchor point group to be used and the first position representation data of all anchor points; and if there exists in the first planar representation image a vertex enclosing region corresponding to some anchor points in the anchor point group to be used, and the number of the some anchor points is not less than a preset number threshold, construct the first virtual grid based on the some anchor points and the first position representation data of the some anchor points;

[0215] The second construction subunit is used to, for any anchor point group to be used, if there is a vertex enclosing area corresponding to each anchor point in the anchor point group to be used in the second plane representation image, then construct the second virtual grid based on all anchor points in the anchor point group to be used and the second position representation data of all anchor points; if there is a vertex enclosing area corresponding to some anchor points in the anchor point group to be used in the second plane representation image, and the number of the some anchor points is not less than a preset number threshold, then construct the second virtual grid based on the some anchor points and the second position representation data of the some anchor points.

[0216] In one possible implementation, the anchor point group to be used includes four anchor points;

[0217] The first construction subunit is specifically configured to: construct two triangular faces in the first virtual grid according to the four anchor points in the anchor point group to be used and the first position representation data of the four anchor points;

[0218] The second construction subunit is specifically configured to construct two triangular faces in the second virtual mesh according to the four anchor points in the anchor point group to be used and the second position representation data of the four anchor points.

[0219] In one possible implementation, the number of the partial anchor points is three;

[0220] The first construction subunit is specifically configured to: construct a triangular face in the first virtual mesh according to the partial anchor points and the first position representation data of the partial anchor points;

[0221] The second construction subunit is specifically configured to construct a triangular face in the second virtual mesh according to the partial anchor points and the second position representation data of the partial anchor points.

[0222] In one possible implementation, the mesh difference representation data includes vertex difference representation data of a plurality of vertices; for any vertex, the vertex difference representation data of the vertex is determined based on difference representation data between the first description information of the vertex and the second description information of the vertex;

[0223] The mesh performance evaluation result is determined based on the Laplace losses corresponding to the several vertices. For any vertex, the Laplace loss corresponding to the vertex is determined based on the vertex difference characterization data of the vertex, the vertex difference characterization data of the adjacent vertices of the vertex, and the two diagonals corresponding to the edge formed by the vertex and the adjacent vertices of the vertex.

[0224] In one possible implementation, the at least one anchor point set includes N anchor point sets arranged sequentially with equal distances between rows and columns, where N is a positive integer;

[0225] The data processing device 900 further includes:

[0226] An anchor point set acquisition unit is configured to determine a starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space according to preset sampling parameters and an arrangement sequence number corresponding to the nth anchor point set; the preset sampling parameters include row spacing, column spacing, and the number of anchor point sets; determine at least one non-starting sampling position corresponding to the nth anchor point set based on the starting sampling position and the preset sampling parameters; and obtain the nth anchor point set by aggregating the anchor points at the starting sampling position and the anchor points at each of the non-starting sampling positions in the preset two-dimensional plane space; n is a positive integer, and n≤N.

[0227] In a possible implementation, the vertex enclosed area is a triangular face; the area description information includes an identifier of the triangular face and at least one of the planar positions of each vertex in the triangular face in the preset two-dimensional plane space.

[0228] In a possible implementation manner, the three-dimensional mesh to be processed is constructed using a three-dimensional reconstruction model;

[0229] The data processing device 900 further includes:

[0230] A model updating unit is used to update the three-dimensional reconstruction model according to the grid performance characterization data.

[0231] Based on the relevant contents of the above-mentioned data processing device 900, it can be known that for the data processing device 900 provided by the embodiment of the present disclosure, after obtaining the three-dimensional mesh to be processed (for example, the reconstructed three-dimensional mesh) and the reference three-dimensional mesh (for example, the pre-set normalized three-dimensional mesh template), a first plane representation image of the three-dimensional mesh to be processed and a second plane representation image of the reference three-dimensional mesh are drawn in a preset two-dimensional plane space (for example, a plane blank image), so that the first plane representation image is used to represent the state of the vertices in the three-dimensional mesh to be processed in the preset two-dimensional plane space (for example, the position distribution state of the vertices, and the connection state between different vertices, etc.), and the second plane representation image is used to represent the state of the vertices in the reference three-dimensional mesh in the preset two-dimensional plane space, so that when the preset two When the three-dimensional plane space includes at least one anchor point set, the first position representation data and the second position representation data of each anchor point in the anchor point set are determined, wherein the first position representation data is determined based on the area description information of the vertex enclosing area corresponding to the anchor point in the first plane representation image, so that the first position representation data can, to a certain extent, represent the state of the three-dimensional mesh to be processed on the anchor point; the second position representation data is determined based on the area description information of the vertex enclosing area corresponding to the anchor point in the second plane representation image, so that the second position representation data can, to a certain extent, represent the state of the reference three-dimensional mesh on the anchor point; then, based on the difference representation data between the first position representation data and the second position representation data, the mesh performance representation data of the three-dimensional mesh to be processed is determined. Among them, because the difference characterization data can represent the difference between the three-dimensional grid to be processed and the reference three-dimensional grid, the grid performance characterization data determined based on the difference characterization data can represent the grid performance presented by the three-dimensional grid to be processed relative to the reference three-dimensional grid, so that the grid performance characterization data can represent the grid performance quantification results presented by the three-dimensional grid to be processed under one or more grid performance evaluation dimensions (for example, dimensions such as standardization). In this way, the purpose of quantitative evaluation processing of a three-dimensional grid under these grid performance evaluation dimensions can be achieved, thereby effectively avoiding defects (for example, poor grid performance evaluation effect, etc.) caused by the inability to quantify a certain grid performance evaluation dimension (for example, dimensions such as standardization) or the poor quantification scheme for a certain grid performance evaluation dimension (for example, dimensions such as accuracy), which is conducive to improving the grid performance evaluation effect.

[0232] In addition, an embodiment of the present disclosure also provides an electronic device, which includes a processor and a memory: the memory is used to store instructions or computer programs; the processor is used to execute the instructions or computer programs in the memory, so that the electronic device executes any implementation of the data processing method provided by the embodiment of the present disclosure.

[0233] Referring to FIG10 , a schematic diagram of the structure of an electronic device 1000 suitable for implementing an embodiment of the present disclosure is shown. The terminal devices in the 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), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The electronic device shown in FIG10 is merely an example and should not limit the functionality and scope of use of the embodiments of the present disclosure.

[0234] As shown in FIG10 , the electronic device 1000 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1008 into a random access memory (RAM) 1003. Various programs and data required for the operation of the electronic device 1000 are also stored in the RAM 1003. The processing device 1001, the ROM 1002, and the RAM 1003 are connected to each other via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0235] Typically, the following devices may be connected to the I / O interface 1005: an input device 1006 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; an output device 1007 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; a storage device 1008 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 may allow the electronic device 1000 to communicate with other devices wirelessly or by wire to exchange data. Although FIG. 10 illustrates the electronic device 1000 with various devices, it should be understood that not all of the devices shown are required to be implemented or present. More or fewer devices may alternatively be implemented or present.

[0236] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a non-transitory computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication device 1009, or installed from the storage device 1008, or installed from the ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment of the present disclosure are performed.

[0237] The electronic device provided by the embodiment of the present disclosure and the method provided by the above embodiment belong to the same inventive concept. For technical details not fully described in this embodiment, please refer to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0238] The embodiments of the present disclosure further provide a computer-readable medium, in which instructions or computer programs are stored. When the instructions or computer programs are executed on a device, the device executes any implementation of the data processing method provided in the embodiments of the present disclosure.

[0239] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, 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 the present disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or component. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a 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. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport 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 thereof.

[0240] In some embodiments, the client and server can communicate using any currently known or later developed network protocol, such as HTTP (Hypertext Transfer Protocol), and can 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 later developed network.

[0241] The computer-readable medium may be included in the electronic device, or may exist independently without being incorporated into the electronic device.

[0242] The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device can perform the method.

[0243] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including, but not limited to, 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 stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through 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., through the Internet using an Internet service provider).

[0244] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two 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 box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0245] The units involved in the embodiments described in this disclosure may be implemented in software or hardware, wherein the name of a unit / module does not, in some cases, limit the unit itself.

[0246] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and 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 chip (SOCs), complex programmable logic devices (CPLDs), and the like.

[0247] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, 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 foregoing.

[0248] It should be noted that the various embodiments of this disclosure are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the descriptions of the systems or devices disclosed in the embodiments for similarities and differences between them. Since the systems or devices disclosed in the embodiments correspond to the methods disclosed in the embodiments, their descriptions are relatively simple, and reference can be made to the descriptions of the methods for any related details.

[0249] It should be understood that in the present disclosure, "at least one (item)" refers to one or more, and "plurality" refers to two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0250] It should also be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0251] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0252] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present disclosure. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present disclosure. Therefore, the present disclosure is not limited to the embodiments shown herein, but is intended to be construed in the widest manner consistent with the principles and novel features disclosed herein.

Claims

1. A data processing method, comprising: Obtaining a three-dimensional mesh to be processed and a reference three-dimensional mesh; Drawing a first plane representation image of the three-dimensional grid to be processed and a second plane representation image of the reference three-dimensional grid in a preset two-dimensional plane space; The first plane representation image is used to represent the state of the vertices in the three-dimensional mesh to be processed in the preset two-dimensional plane space; The second plane representation image is used to represent the state of the vertices in the reference three-dimensional mesh in the preset two-dimensional plane space; the preset two-dimensional plane space includes at least one anchor point set; Determine first position representation data and second position representation data of an anchor point in each of the anchor point sets; the first position representation data is determined based on region description information of a vertex enclosed region corresponding to the anchor point in the first plane representation image; the second position representation data is determined based on region description information of a vertex enclosed region corresponding to the anchor point in the second plane representation image; The grid performance characterization data of the three-dimensional grid to be processed is determined according to the difference characterization data between the first position characterization data and the second position characterization data.

2. The method according to claim 1, wherein: The process of determining the grid performance characterization data includes: For any anchor point set, a first virtual grid and a second virtual grid are constructed using the anchor point set, and a grid performance evaluation result corresponding to the anchor point set is determined based on the grid difference representation data between the first virtual grid and the second virtual grid; the vertices in the first virtual grid are determined based on the first anchor point in the anchor point set, a vertex enclosing area corresponding to the first anchor point exists in the first plane representation image, and the first description information of the vertices in the first virtual grid is determined based on the first position representation data of the first anchor point; the vertices in the second virtual grid are determined based on the second anchor point in the anchor point set, a vertex enclosing area corresponding to the second anchor point exists in the second plane representation image, and the second description information of the vertices in the second virtual grid is determined based on the second position representation data of the second anchor point; the grid difference representation data is determined based on the difference representation data between the first description information of the vertices in the first virtual grid and the second description information of the vertices in the second virtual grid; Determine mesh performance characterization data of the three-dimensional mesh to be processed according to the mesh performance evaluation result corresponding to the at least one anchor point set.

3. The method according to claim 2, wherein: The method of constructing the first virtual grid and the second virtual grid by using the anchor point set includes: Determine each anchor point group to be used from the anchor point set; for any anchor point group to be used, a rectangle formed by all anchor points in the anchor point group to be used satisfies a preset rectangle condition; For any anchor point group to be used, if there is a vertex enclosed area corresponding to each anchor point in the anchor point group to be used in the first plane representation image, then according to all anchor points in the anchor point group to be used and the first position representation data of all anchor points constructing the first virtual grid; if there are vertex enclosed areas corresponding to some anchor points in the to-be-used anchor point group in the first plane representation image, and the number of the some anchor points is not less than a preset number threshold, constructing the first virtual grid according to the some anchor points and the first position representation data of the some anchor points; For any anchor point group to be used, if there exists in the second plane representation image a vertex enclosing area corresponding to each anchor point in the anchor point group to be used, the second virtual grid is constructed based on all the anchor points in the anchor point group to be used and the second position representation data of all the anchor points; if there exists in the second plane representation image a vertex enclosing area corresponding to some of the anchor points in the anchor point group to be used, and the number of the some anchor points is not less than a preset number threshold, the second virtual grid is constructed based on the some anchor points and the second position representation data of the some anchor points.

4. The method according to claim 3, wherein: The anchor point group to be used includes four anchor points; The step of constructing the first virtual grid according to all anchor points in the anchor point group to be used and the first position representation data of all anchor points includes: Constructing two triangular faces in the first virtual grid according to four anchor points in the anchor point group to be used and first position representation data of the four anchor points; The step of constructing the second virtual grid according to all anchor points in the anchor point group to be used and the second position representation data of all anchor points includes: Two triangular faces in the second virtual mesh are constructed according to the four anchor points in the anchor point group to be used and the second position representation data of the four anchor points.

5. The method according to claim 3, wherein: The number of the partial anchor points is three; The constructing the first virtual grid according to the part of anchor points and the first position representation data of the part of anchor points includes: Constructing a triangular face in the first virtual grid according to the partial anchor points and the first position representation data of the partial anchor points; The constructing the second virtual grid according to the part of the anchor points and the second position representation data of the part of the anchor points includes: A triangular face in the second virtual mesh is constructed according to the partial anchor points and the second position representation data of the partial anchor points.

6. The method according to claim 2, wherein: The mesh difference representation data includes vertex difference representation data of a plurality of vertices; for any vertex, the vertex difference representation data of the vertex is determined based on the difference representation data between the first description information of the vertex and the second description information of the vertex; The mesh performance evaluation result is determined based on the Laplace losses corresponding to the several vertices. For any vertex, the Laplace loss corresponding to the vertex is determined based on the vertex difference characterization data of the vertex, the vertex difference characterization data of the adjacent vertices of the vertex, and the two diagonals corresponding to the edges formed by the vertex and the adjacent vertices of the vertex.

7. The method according to claim 1, wherein: The at least one anchor point set comprises N anchor point sets arranged in sequence with equal distances between rows and columns, where N is a positive integer; The process of obtaining the nth anchor point set includes: According to preset sampling parameters and the arrangement sequence number corresponding to the nth anchor point set, determine the starting sampling position corresponding to the nth anchor point set in the preset two-dimensional plane space; the preset sampling parameters include row spacing, column spacing and the number of anchor point sets; n is a positive integer, n≤N; Determining at least one non-starting sampling position corresponding to the nth anchor point set according to the starting sampling position and the preset sampling parameters; The anchor points at the starting sampling position and the anchor points at each of the non-starting sampling positions in the preset two-dimensional plane space are grouped together to obtain the nth anchor point set.

8. The method according to claim 1, wherein: The vertex enclosed area is a triangular surface; The region description information includes at least one of an identifier of the triangular face and a plane position of each vertex in the triangular face in the preset two-dimensional plane space.

9. The method according to claim 1, wherein: The three-dimensional mesh to be processed is constructed using a three-dimensional reconstruction model; After determining the mesh performance characterization data of the three-dimensional mesh to be processed, the method further includes: The three-dimensional reconstruction model is updated according to the grid performance characterization data.

10. A data processing device, comprising: A mesh acquisition unit, configured to acquire a three-dimensional mesh to be processed and a reference three-dimensional mesh; An image drawing unit is configured to draw a first plane representation image of the three-dimensional grid to be processed and a second plane representation image of the reference three-dimensional grid in a preset two-dimensional plane space; The first plane representation image is used to represent the state of the vertices in the three-dimensional mesh to be processed in the preset two-dimensional plane space; The second plane representation image is used to represent the state of the vertices in the reference three-dimensional mesh in the preset two-dimensional plane space; the preset two-dimensional plane space includes at least one anchor point set; A first determining unit is configured to determine first position representation data and second position representation data of an anchor point in each of the anchor point sets; the first position representation data is determined based on region description information of a vertex enclosed region corresponding to the anchor point in the first plane representation image; the second position representation data is determined based on region description information of a vertex enclosed region corresponding to the anchor point in the second plane representation image; The second determining unit is configured to determine the mesh performance characterization data of the to-be-processed three-dimensional mesh according to the difference characterization data between the first position characterization data and the second position characterization data.

11. An electronic device comprising a processor and a memory, wherein: The memory is configured to store instructions or computer programs; The processor is configured to execute the instructions or computer programs in the memory so that the electronic device performs the method according to any one of claims 1 to 9.

12. A computer readable medium storing instructions or a computer program, wherein: When the instructions or computer programs are executed on a device, the device is caused to execute the method according to any one of claims 1 to 9.

Images